Intelligent Blood Pressure Measurement Method and System Based on Electronic Korotkoff Sound Method

Abstract

The present application provides an intelligent blood pressure measurement method and system based on the electronic Korotkoff sound method, which relates to the technical field of medical devices. The method includes: during the process of deflating the cuff wrapped around the upper arm of the person to be measured, obtaining a plurality of Korotkoff sound signals from the brachial artery and the air pressure in the cuff corresponding to each Korotkoff sound signal respectively; determining the initial systolic blood pressure of the person to be measured according to the first air pressure in the cuff when the Korotkoff sound signal first appears, and determining the initial diastolic blood pressure of the person to be measured according to the second air pressure in the cuff when the Korotkoff sound signal appears last; extracting eigenvalue from the plurality of Korotkoff sound signals to obtain a plurality of characteristic parameters; inputting the plurality of characteristic parameters, the initial systolic blood pressure and the initial diastolic blood pressure into a preset neural network model to obtain the target systolic blood pressure and the target diastolic blood pressure of the person to be measured, and taking the target systolic blood pressure and the target diastolic blood pressure as the blood pressure measurement result of the person to be measured. The technical effect of the present application is: improving the accuracy of blood pressure measurement.

Description

Technical Field

[0001] This application relates to the technical field of medical devices, and particularly to an intelligent blood pressure measurement method and system based on the electronic Korotkoff sound method. Background Art

[0002] Blood pressure, as one of the important physiological indicators of the human body, the accuracy of its measurement is directly related to the diagnosis and treatment effect of diseases. At present, electronic sphygmomanometers based on the Korotkoff sound method are widely used because of their simple operation and low cost. However, during the measurement process, they are easily affected by various factors such as environmental noise, human activities, and measurement postures, resulting in difficulty in ensuring the accuracy and stability of the measurement results.

[0003] In the prior art, a method of setting a fixed threshold is usually adopted to identify Korotkoff sound signals and determine blood pressure values, that is, when the detected signal amplitude exceeds the preset threshold, it is determined as a valid Korotkoff sound signal, and the cuff pressures corresponding to the first appearance and disappearance are respectively used as the systolic blood pressure and diastolic blood pressure. However, due to the differences in physiological characteristics of different individuals, the intensities and characteristics of Korotkoff sound signals are also different. It is difficult to adapt to this individual difference with a unified fixed threshold, which easily causes blood pressure measurement deviation. Summary of the Invention

[0004] This application provides an intelligent blood pressure measurement method based on the electronic Korotkoff sound method to improve the accuracy of blood pressure measurement.

[0005] In a first aspect, this application provides an intelligent blood pressure measurement method based on the electronic Korotkoff sound method. The method includes: after increasing the air pressure in the cuff wound around the upper arm of the person to be measured to a preset pressure, deflating the cuff; during the process of deflating the cuff, obtaining a plurality of Korotkoff sound signals from the brachial artery, and the air pressure in the cuff corresponding to each of the Korotkoff sound signals; determining the initial systolic blood pressure of the person to be measured according to the first air pressure in the cuff when the Korotkoff sound signal first appears among the plurality of Korotkoff sound signals, and determining the initial diastolic blood pressure of the person to be measured according to the second air pressure in the cuff when the Korotkoff sound signal appears last among the plurality of Korotkoff sound signals; extracting eigenvalue of the plurality of Korotkoff sound signals to obtain a plurality of characteristic parameters; inputting the plurality of characteristic parameters, the initial systolic blood pressure, and the initial diastolic blood pressure into a preset neural network model to obtain the target systolic blood pressure and target diastolic blood pressure of the person to be measured, and taking the target systolic blood pressure and the target diastolic blood pressure as the blood pressure measurement result of the person to be measured.

[0006] By adopting the above technical solution, by obtaining a plurality of Korotkoff sound signals and corresponding cuff pressures during the deflation process, not only can the initial systolic blood pressure and initial diastolic blood pressure be determined according to the first and last occurrence times of the Korotkoff sound signals, but also the obtained Korotkoff sound signals are subjected to deep feature extraction to obtain a plurality of feature parameters including frequency domain eigenvalue, amplitude eigenvalue and time domain eigenvalue. These extracted feature parameters, together with the initial systolic blood pressure and initial diastolic blood pressure, are input into a preset neural network model. Through the deep learning ability of the model, the rich feature information in the Korotkoff sound signals is fully exploited, realizing an accurate mapping from the multi-dimensional feature space to the blood pressure value. This method not only overcomes the problem of poor adaptability of the traditional fixed threshold method to individual differences, but also can adaptively learn the signal features under different measurement conditions, improving the accuracy of blood pressure measurement.

[0007] Optionally, the obtaining of a plurality of Korotkoff sound signals from the brachial artery and the cuff pressures respectively corresponding to the Korotkoff sound signals includes: during the deflation of the cuff, receiving the plurality of Korotkoff sound signals generated by the pulse beating of the brachial artery collected by a sound sensor disposed in the cuff, and receiving the cuff pressures respectively corresponding to the Korotkoff sound signals sent by a pressure sensor disposed in the cuff.

[0008] By adopting the above technical solution, by disposing a sound sensor and a pressure sensor in the cuff, the synchronous acquisition of the Korotkoff sound signals and the corresponding cuff pressures is realized. The sound sensor can accurately capture the Korotkoff sound signals generated by the pulse beating of the brachial artery, while the pressure sensor monitors the air pressure change in the cuff in real time. The combination of the two ensures the timing of signal acquisition and the accuracy of the corresponding relationship, providing a reliable raw data basis for subsequent feature extraction and blood pressure calculation.

[0009] Optionally, the eigenvalue extraction of the plurality of Korotkoff sound signals to obtain a plurality of feature parameters includes: performing Fourier transform on each of the Korotkoff sound signals to obtain the frequency domain eigenvalue of each of the Korotkoff sound signals; performing amplitude normalization processing on each of the Korotkoff sound signals to obtain the amplitude eigenvalue of each of the Korotkoff sound signals; obtaining the time domain eigenvalue of each of the Korotkoff sound signals according to the duration of each of the Korotkoff sound signals; and using the frequency domain eigenvalue, the amplitude eigenvalue and the time domain eigenvalue as the plurality of feature parameters of each of the Korotkoff sound signals.

[0010] By adopting the above technical solution, through performing Fourier transform, amplitude normalization processing and time feature analysis on the Korotkoff sound signal respectively, the frequency domain eigenvalue, amplitude eigenvalue and time domain eigenvalue of the signal are obtained, realizing multi-dimensional feature extraction of the Korotkoff sound signal. This multi-dimensional feature extraction method can comprehensively capture the frequency distribution, amplitude change and timing characteristics of the signal, enabling the extracted feature parameters to more completely represent the characteristics of the Korotkoff sound signal, providing rich input information for the neural network model, and thus improving the accuracy of blood pressure measurement.

[0011] Optionally, before inputting the multiple feature parameters, the initial systolic blood pressure and the initial diastolic blood pressure into the preset neural network model, it further includes: obtaining multiple groups of sample data, each group of sample data including the feature parameters of multiple Korotkoff sound signals, the initial systolic blood pressure, the initial diastolic blood pressure, and the standard systolic blood pressure and standard diastolic blood pressure measured by a standard blood pressure measurement device; using the feature parameters, the initial systolic blood pressure and the initial diastolic blood pressure in each group of sample data as input features, and using the standard systolic blood pressure and the standard diastolic blood pressure as output features; training the neural network structure according to the input features and the output features to obtain the preset neural network model.

[0012] By adopting the above technical solution, by collecting a large number of sample data for neural network training, where each group of sample data contains complete feature parameters, initial blood pressure values and standard blood pressure values, enabling the neural network to learn the mapping relationship between the feature parameters and the actual blood pressure values. This training method calibrated based on a standard blood pressure measurement device enables the preset neural network model to have the ability to convert multi-dimensional features into accurate blood pressure values, effectively improving the generalization performance and prediction accuracy of the model, and ensuring the consistency of the final blood pressure measurement result with the standard device.

[0013] Optionally, after obtaining the preset neural network model, it further includes: obtaining multiple groups of verification sample data, each group of verification sample data including the verification feature parameters of multiple Korotkoff sound signals, the verification initial systolic blood pressure, the verification initial diastolic blood pressure, and the verification standard systolic blood pressure and verification standard diastolic blood pressure measured by a standard blood pressure measurement device; inputting the verification feature parameters, the verification initial systolic blood pressure and the verification initial diastolic blood pressure in each group of verification sample data into the preset neural network model to obtain the predicted systolic blood pressure and the predicted diastolic blood pressure; calculating the first difference between the predicted systolic blood pressure and the verification standard systolic blood pressure, and the second difference between the predicted diastolic blood pressure and the verification standard diastolic blood pressure; when both the first difference and the second difference meet the preset requirements, determining that the preset neural network model meets the usage requirements, and if there is the first difference and / or the second difference that does not meet the preset requirements, retraining the neural network model until a preset neural network model that meets the usage requirements is obtained.

[0014] By adopting the above technical solution, a performance verification of the trained neural network model is carried out by introducing independent verification sample data, and the accuracy of the model is evaluated by calculating the difference between the predicted value and the standard value, thereby establishing a strict model quality control mechanism. When the differences between the predicted systolic blood pressure and diastolic blood pressure and the verification standard values do not meet the preset requirements, the model retraining process is triggered. This iterative optimization mechanism ensures that the finally put-into-use neural network model has reliable prediction accuracy, significantly improving the accuracy and credibility of blood pressure measurement results.

[0015] Optionally, the step of inputting the multiple feature parameters, the initial systolic blood pressure and the initial diastolic blood pressure into a preset neural network model to obtain the target systolic blood pressure and target diastolic blood pressure of the person to be measured includes: inputting the multiple feature parameters, the initial systolic blood pressure and the initial diastolic blood pressure into the preset neural network model through the input layer of the preset neural network model; using the hidden layer of the preset neural network model to perform feature extraction and non-linear transformation on the input multiple feature parameters, the initial systolic blood pressure and the initial diastolic blood pressure; and obtaining the target systolic blood pressure and target diastolic blood pressure of the person to be measured through the output layer of the preset neural network model.

[0016] By adopting the above technical solution, through the three-layer structure design of the neural network model, an efficient conversion process from the original features to the blood pressure values is realized. The input layer receives multi-dimensional feature parameters and initial blood pressure values, the hidden layer fully explores the deep associations between the features through feature extraction and non-linear transformation, and the output layer finally generates the target systolic blood pressure and diastolic blood pressure. This hierarchical data processing mechanism can effectively capture the complex mapping relationship between the feature parameters and the blood pressure values, improving the accuracy and stability of blood pressure prediction.

[0017] Optionally, after taking the target systolic blood pressure and the target diastolic blood pressure as the blood pressure measurement results of the person to be measured, the method further includes: comparing the target systolic blood pressure and the target diastolic blood pressure with a preset normal range of systolic blood pressure and a preset normal range of diastolic blood pressure respectively; when the target systolic blood pressure exceeds the preset normal range of systolic blood pressure and / or the target diastolic blood pressure exceeds the preset normal range of diastolic blood pressure, generating a blood pressure abnormality prompt message; and displaying the target systolic blood pressure, the target diastolic blood pressure and the blood pressure abnormality prompt message on a display screen.

[0018] By adopting the above technical solution, the measured target systolic blood pressure and target diastolic blood pressure are compared with the preset normal range in real time. When blood pressure abnormalities are detected, prompt messages are automatically generated, and the blood pressure measurement results and abnormal prompt messages are presented on the display screen simultaneously, realizing the intelligent monitoring and warning functions of blood pressure measurement results. This timely abnormal prompt mechanism can help users quickly identify blood pressure abnormalities, contribute to users taking corresponding health management measures in a timely manner, and improve the practical value of blood pressure measurement and the effect of health monitoring.

[0019] In a second aspect, the present application provides an intelligent blood pressure measurement system based on the electronic Korotkoff sound method. The system includes: a control module, an acquisition module, a determination module, an extraction module, and an output module. Among them, the control module is used to increase the air pressure in the cuff wrapped around the upper arm of the person to be measured to a preset pressure and then deflate the cuff. The acquisition module is used to acquire a plurality of Korotkoff sound signals from the brachial artery and the air pressure in the cuff corresponding to each of the Korotkoff sound signals during the deflation of the cuff. The determination module is used to determine the initial systolic blood pressure of the person to be measured according to the first air pressure in the cuff when the Korotkoff sound signal first appears among the plurality of Korotkoff sound signals, and determine the initial diastolic blood pressure of the person to be measured according to the second air pressure in the cuff when the Korotkoff sound signal appears last among the plurality of Korotkoff sound signals. The extraction module is used to extract eigenvalue of the plurality of Korotkoff sound signals to obtain a plurality of characteristic parameters. The output module is used to input the plurality of characteristic parameters, the initial systolic blood pressure, and the initial diastolic blood pressure into a preset neural network model to obtain the target systolic blood pressure and target diastolic blood pressure of the person to be measured, and use the target systolic blood pressure and the target diastolic blood pressure as the blood pressure measurement results of the person to be measured.

[0020] In a third aspect, the present application provides an electronic device, adopting the following technical solution: including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions. The user interface and the network interface are used to communicate with other devices. The processor is used to execute the instructions stored in the memory to enable the electronic device to execute a computer program of any one of the above intelligent blood pressure measurement methods based on the electronic Korotkoff sound method.

[0021] In a fourth aspect, the present application provides a computer-readable storage medium, adopting the following technical solution: storing a computer program that can be loaded and executed by a processor for any one of the above intelligent blood pressure measurement methods based on the electronic Korotkoff sound method.

[0022] In summary, the present application includes at least one of the following beneficial technical effects:

[0023] 1. By obtaining multiple Korotkoff sound signals and corresponding cuff pressures during the deflation process, not only can the initial systolic blood pressure and initial diastolic blood pressure be determined according to the first and last occurrence times of the Korotkoff sound signals, but also deep feature extraction is performed on the obtained Korotkoff sound signals to obtain multiple characteristic parameters including frequency domain characteristic values, amplitude characteristic values, and time domain characteristic values. These extracted characteristic parameters, together with the initial systolic blood pressure and initial diastolic blood pressure, are input into a preset neural network model. Through the deep learning ability of the model, the rich characteristic information in the Korotkoff sound signals is fully exploited, realizing an accurate mapping from the multi-dimensional feature space to the blood pressure value. This method not only overcomes the problem of poor adaptability of the traditional fixed threshold method to individual differences, but also can adaptively learn the signal characteristics under different measurement conditions, improving the accuracy of blood pressure measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a schematic flowchart of an intelligent blood pressure measurement method based on the electronic Korotkoff sound method provided by an embodiment of the present application;

[0025] Figure 2 is a schematic structural diagram of an intelligent blood pressure measurement device based on the electronic Korotkoff sound method provided by an embodiment of the present application;

[0026] Figure 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

[0027] Description of the reference numerals: 1000, electronic device; 1001, processor; 1002, communication bus; 1003, user interface; 1004, network interface; 1005, memory. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments.

[0029] In the description of the embodiments of the present application, words such as "exemplary", "for example", or "for illustration" are used to represent examples, illustrations, or explanations. Any embodiment or design solution described as "exemplary", "for example", or "for illustration" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, the use of words such as "exemplary", "for example", or "for illustration" is intended to present relevant concepts in a specific manner.

[0030] Figure 1 is a schematic flowchart of an intelligent blood pressure measurement method based on the electronic Korotkoff sound method provided by an embodiment of the present application. As Figure 1As shown, the method includes S101 - S105:

[0031] S101, after increasing the air pressure in the cuff wrapped around the upper arm of the person to be measured to a preset pressure, deflate the cuff.

[0032] When specifically implementing the present invention, the pressurization and deflation steps need to be executed first. The purpose of this step is to create necessary conditions for subsequent acquisition of Korotkoff sound signals by controlling the change of air pressure in the cuff.

[0033] Specifically, wrap the cuff of the electronic sphygmomanometer around the upper arm position of the person to be measured, align the center of the cuff with the brachial artery. By starting the air pump, quickly increase the air pressure in the cuff to the preset pressure, which is usually set to 180 mmHg, or increase by 20 - 30 mmHg according to the systolic blood pressure value measured for the person to be measured last time. When the air pressure in the cuff reaches the preset value, at this time the brachial artery is completely compressed and blood flow is temporarily blocked. Subsequently, control the solenoid valve to deflate the cuff. It provides a stable pressure change environment for subsequent acquisition of Korotkoff sound signals, which helps to improve the accuracy of blood pressure measurement.

[0034] S102, during the process of deflating the cuff, acquire multiple Korotkoff sound signals from the brachial artery and the air pressure in the cuff corresponding to each Korotkoff sound signal respectively.

[0035] Korotkoff sound is a characteristic sound generated by blood flow during blood pressure measurement.

[0036] When the cuff pressure is higher than the systolic blood pressure, the brachial artery is completely closed, and at this time there is no blood flow passing through, so no sound can be heard. As the cuff pressure gradually decreases and is lower than the systolic blood pressure, every time the heart contracts, a small amount of blood will pass through the partially compressed brachial artery in a turbulent manner, and this turbulence will cause the blood vessel wall to vibrate, thus generating a sound that can be detected, namely Korotkoff sound.

[0037] The five characteristic stages of Korotkoff sound: The first stage (appearance stage): When the cuff pressure is slightly lower than the systolic blood pressure, a clear tapping sound appears for the first time, and this pressure value corresponds to the systolic blood pressure; The second stage (enhancement stage): The sound becomes louder and longer, with a swishing sound; The third stage (loud stage): The sound reaches the maximum intensity;

[0038] The fourth stage (weakening stage): The sound suddenly becomes dull and soft; The fifth stage (disappearance stage): The sound completely disappears, and the pressure value at this time corresponds to the diastolic blood pressure.

[0039] Characteristics of Korotkoff sound signals: Frequency characteristics: mainly concentrated in the range of 20 - 200 Hz; Amplitude characteristics: the sound intensity first increases and then decreases with the change of cuff pressure; Time-domain characteristics: the duration and occurrence time of Korotkoff sound within each pulse cycle.

[0040] Specifically, during the process of cuff deflation, it is necessary to synchronously collect Korotkoff sound signals and the corresponding air pressure values inside the cuff. This is because there is a direct corresponding relationship between the appearance and disappearance of Korotkoff sound signals and the air pressure inside the cuff. Through this corresponding relationship, the blood pressure value can be accurately judged.

[0041] In specific implementation, a sound sensor and a pressure sensor are arranged inside the cuff. Among them, the sound sensor is preferably a high-sensitivity electret microphone for collecting Korotkoff sound signals from the brachial artery; the pressure sensor is a piezoresistive sensor with an accuracy of 0.1 mmHg for real-time monitoring of the air pressure change inside the cuff. When the air pressure inside the cuff gradually decreases, the sound sensor starts to capture Korotkoff sound signals generated when the blood flow in the brachial artery passes through, and at the same time, the pressure sensor continuously collects the air pressure value inside the cuff at this time. The sampling frequency is set to 1 kHz to ensure that the subtle changes of Korotkoff sound signals can be accurately captured. The collected Korotkoff sound signals are amplified by a preamplifier and filtered by a band-pass filter (20 Hz - 200 Hz) to remove ambient noise and other interference signals. Through the microprocessor, the collected Korotkoff sound signals are paired and stored with the corresponding air pressure values inside the cuff at the corresponding moments to form time-series data. This synchronous acquisition method can not only accurately record the appearance and disappearance moments of Korotkoff sound signals, but also ensure that each Korotkoff sound signal has its precise corresponding air pressure value inside the cuff, providing a reliable data basis for subsequent blood pressure calculation.

[0042] Based on the above embodiments, as an alternative implementation manner, in S102, obtaining multiple Korotkoff sound signals from the brachial artery and the corresponding air pressure inside the cuff for each Korotkoff sound signal specifically includes:

[0043] During the process of deflating the cuff, receive multiple Korotkoff sound signals generated by the pulse beating of the brachial artery collected by the sound sensor arranged inside the cuff, and receive the air pressure inside the cuff corresponding to each Korotkoff sound signal sent by the pressure sensor arranged inside the cuff.

[0044] S103, determine the initial systolic blood pressure of the person to be measured according to the first air pressure inside the cuff when the Korotkoff sound signal first appears among the multiple Korotkoff sound signals, and determine the initial diastolic blood pressure of the person to be measured according to the second air pressure inside the cuff when the Korotkoff sound signal appears last among the multiple Korotkoff sound signals.

[0045] In one example, after obtaining a complete sequence of Korotkoff sound signals, it is necessary to determine the initial systolic blood pressure and the initial diastolic blood pressure of the person to be measured, which are the basic steps of blood pressure measurement.

[0046] First, perform signal processing on the collected Korotkoff sound signals, including band-pass filtering and envelope extraction of the signals, to highlight the characteristics of the Korotkoff sound and eliminate interference. By setting an appropriate amplitude threshold (e.g., 3 times the mean of the background noise), the appearance and disappearance moments of the Korotkoff sound signals can be accurately identified. When the first Korotkoff sound signal exceeding this threshold is detected, record the first air pressure in the cuff at this moment, and use it as the initial systolic blood pressure of the person to be measured; when the last Korotkoff sound signal exceeding this threshold is detected, record the second air pressure in the cuff at this moment, and use it as the initial diastolic blood pressure of the person to be measured.

[0047] To improve the reliability of the measurement, a verification mechanism of three consecutive cardiac cycles is adopted during the identification process, that is, only when Korotkoff sound signals are detected in three consecutive cycles can it be confirmed as a valid signal, which can effectively avoid misjudgment caused by accidental noise. This method for determining the initial blood pressure based on the characteristic points of the Korotkoff sound signal can provide reliable initial input parameters for the subsequent neural network model, and at the same time retains the basic principle of the traditional Korotkoff sound method measurement. The accurate determination of the initial systolic blood pressure and the initial diastolic blood pressure lays a foundation for obtaining more accurate blood pressure measurement results through the subsequent neural network model.

[0048] S104, extract eigenvalue from multiple Korotkoff sound signals to obtain multiple characteristic parameters.

[0049] Specifically, in order to make full use of the blood pressure characteristic information contained in the Korotkoff sound signals, it is necessary to comprehensively extract eigenvalues from the obtained multiple Korotkoff sound signals. Extract the time-domain, frequency-domain, and amplitude characteristics of each Korotkoff sound signal respectively to obtain a complete set of characteristic parameters.

[0050] First, perform Fourier transform on the Korotkoff sound signal, calculate the spectral characteristics of the signal, extract the energy distribution of the main frequency components (in the range of 20 - 200 Hz), and obtain the frequency domain eigenvalue, including parameters such as the main frequency of the signal, the bandwidth, and the energy ratio of each frequency band. Secondly, perform amplitude normalization on the Korotkoff sound signal, map the signal amplitude to a unified range (such as between 0 and 1), and extract the amplitude eigenvalue, including parameters such as the maximum amplitude, average amplitude, and amplitude variance of the signal. This normalization process can eliminate the differences in signal intensity under different measurement environments. At the same time, analyze the time domain characteristics of the Korotkoff sound signal, measure the time domain eigenvalue such as the duration, rise time, and peak occurrence time of each Korotkoff sound signal. These parameters reflect the vascular elasticity and hemodynamic characteristics. Combine the obtained frequency domain eigenvalues, amplitude eigenvalues, and time domain eigenvalues to form a set of characteristic parameters. These characteristic parameters comprehensively reflect various characteristics of the Korotkoff sound signal, provide rich input features for the subsequent neural network model, and help improve the accuracy of blood pressure measurement. Through this multi-dimensional feature extraction method, not only can the basic features relied on by traditional measurement methods be captured, but also more potential features reflecting blood pressure changes can be obtained, providing more comprehensive data support for intelligent blood pressure measurement.

[0051] Based on the above embodiments, as an alternative implementation, in S104, extracting eigenvalues from multiple Korotkoff sound signals to obtain multiple characteristic parameters specifically includes S401 - S404:

[0052] S401, perform Fourier transform on each Korotkoff sound signal to obtain the frequency domain eigenvalue of each Korotkoff sound signal.

[0053] Specifically, preprocess each collected Korotkoff sound signal, including removing the DC component and windowing. Select the Hanning window function to reduce spectral leakage. Subsequently, perform the Fast Fourier Transform (FFT) on the processed signal to convert the time domain signal into a frequency domain representation.

[0054] Since the effective frequency components of the Korotkoff sound signal are mainly distributed in the range of 20 - 200 Hz, the spectral characteristics in this frequency band are analyzed emphatically. By analyzing the spectrum, the following frequency domain eigenvalues are extracted: the main frequency (the frequency corresponding to the maximum spectral energy), the bandwidth (the frequency range where the spectral energy exceeds 50% of the maximum value), the spectral centroid (reflecting the central position of the spectral energy distribution), the energy ratio of each frequency band (divide 20 - 200 Hz into several sub-bands and calculate the ratio of the energy of each sub-band to the total energy), and the spectral peak ratio (the ratio of the secondary peak to the main peak). These frequency domain eigenvalues reflect the frequency composition characteristics of the Korotkoff sound signal, can reflect the changes in the vascular wall vibration characteristics and blood flow state, and provide important frequency domain information input for the subsequent neural network model.

[0055] S402. Perform amplitude normalization processing on each Korotkoff sound signal to obtain the amplitude eigenvalue of each Korotkoff sound signal.

[0056] Specifically, perform normalization processing on each Korotkoff sound signal. Using the maximum-minimum normalization method, map the signal amplitude to the interval [0, 1]. The calculation formula is: X_normalized = (X - X_min) / (X_max - X_min), where X is the original signal value, and X_min and X_max are the minimum and maximum values of the signal respectively.

[0057] After normalization processing, calculate the following amplitude eigenvalues: signal maximum amplitude (the peak amplitude after normalization), average amplitude (the mean value of the normalized signal), amplitude standard deviation (reflecting the degree of dispersion of the signal amplitude), waveform factor (the ratio of the effective value to the average value), peak factor (the ratio of the maximum value to the effective value), and impulse factor (the ratio of the maximum value to the average value). These amplitude eigenvalues can comprehensively reflect the intensity change characteristics of the Korotkoff sound signal, eliminate the signal amplitude differences caused by different measurement conditions, and make the Korotkoff sound signals obtained at different times and in different measurement environments comparable. Through this standardization processing and feature extraction method, not only the amplitude change characteristics of the signal are retained, but also the stability of the feature parameters is improved, providing reliable amplitude feature inputs for the neural network model, which helps to improve the accuracy and repeatability of blood pressure measurement.

[0058] S403. Obtain the time-domain eigenvalue of each Korotkoff sound signal according to the duration of each Korotkoff sound signal.

[0059] Specifically, set an amplitude threshold (usually 10% of the signal maximum amplitude) to determine the starting point and ending point of each Korotkoff sound signal, so as to calculate the duration of the signal. Based on the determined time range, extract the following time-domain eigenvalues: signal duration (the time interval from the starting point to the ending point), rise time (the time from the starting point to the peak), fall time (the time from the peak to the ending point), peak occurrence time (the time position relative to the signal starting point), zero-crossing rate (the number of times the signal crosses the zero point per unit time), and the first-order difference of the signal (reflecting the signal change rate).

[0060] These time-domain eigenvalues reflect the time characteristics of blood vessel wall vibration and hemodynamic characteristics. In particular, the signal duration and rise time are closely related to blood vessel elasticity and blood pressure level. In actual processing, the sliding window method is used to perform segmented analysis on the signal. The window length is set to one cardiac cycle to ensure the continuity and integrity of feature extraction. By analyzing the change trend of these time-domain eigenvalues, the dynamic change characteristics of the blood vessel state during blood pressure change can be reflected, providing important time-domain information for the neural network model, which helps to improve the accuracy and reliability of blood pressure measurement.

[0061] S404. Use the frequency domain eigenvalue, amplitude eigenvalue, and time domain eigenvalue as multiple characteristic parameters of each Korotkoff sound signal.

[0062] Based on the above embodiments, as an alternative embodiment, before inputting the multiple characteristic parameters, initial systolic blood pressure, and initial diastolic blood pressure into the preset neural network model, it specifically further includes S501 - S503:

[0063] S501. Obtain multiple groups of sample data. Each group of sample data includes characteristic parameters of multiple Korotkoff sound signals, initial systolic blood pressure, initial diastolic blood pressure, and standard systolic blood pressure and standard diastolic blood pressure measured by a standard blood pressure measuring device.

[0064] S502. Use the characteristic parameters, initial systolic blood pressure, and initial diastolic blood pressure in each group of sample data as input characteristics, and use the standard systolic blood pressure and standard diastolic blood pressure as output characteristics.

[0065] Specifically, collect multiple groups of sample data through clinical experiments. The experimental subjects include subjects of different ages, genders, and health conditions. Each subject measures simultaneously using the measurement method of the present invention and a medical mercury sphygmomanometer. For each group of sample data, record the characteristic parameters (including frequency domain eigenvalue, amplitude eigenvalue, and time domain eigenvalue) extracted from the Korotkoff sound signal, as well as the initial systolic blood pressure and initial diastolic blood pressure determined based on the first and last Korotkoff sounds. At the same time, record the standard systolic blood pressure and standard diastolic blood pressure measured by the medical mercury sphygmomanometer as the true values.

[0066] The process of collecting sample data strictly follows medical measurement specifications to ensure the accuracy and reliability of the data. In the data pre - processing stage, normalize all characteristic parameters so that they are distributed within the same numerical range to avoid affecting the model training effect due to too large a difference in numerical ranges. Combine the processed set of characteristic parameters with the initial systolic blood pressure and initial diastolic blood pressure to form an input feature vector, and use the corresponding standard systolic blood pressure and standard diastolic blood pressure as the output feature vector to form a complete training data pair.

[0067] S503. Train the neural network structure according to the input characteristics and output characteristics to obtain the preset neural network model.

[0068] In an example, to build an intelligent model that can accurately predict blood pressure values, it is necessary to train the neural network based on the training data pairs prepared in the foregoing steps.

[0069] The neural network adopts a multi-layer perceptron structure, including an input layer, two hidden layers and an output layer. The number of neurons in the input layer corresponds to the dimension of the feature parameters (including frequency-domain feature values, amplitude feature values, time-domain feature values, as well as the initial systolic blood pressure and the initial diastolic blood pressure). The two hidden layers are respectively configured with 64 and 32 neurons, and the output layer contains 2 neurons corresponding to the target systolic blood pressure and the target diastolic blood pressure.

[0070] The network uses ReLU as the activation function for the hidden layers and a linear activation function for the output layer. The mean squared error (MSE) is selected as the loss function to evaluate the difference between the predicted value and the true value. The mini-batch stochastic gradient descent algorithm is used in the training process. The batch size is set to 64, the initial learning rate is set to 0.001, and the Adam optimizer is used to adaptively adjust the learning rate. To prevent overfitting, a Dropout layer is added between the two hidden layers with a dropout rate of 0.3, and the L2 regularization technique is also used. The overall data is divided into a training set, a validation set and a test set in the ratio of 7:2:1. During the training process, the model performance is monitored through the validation set. When the validation set loss has not improved for 5 consecutive epochs, the early stopping mechanism is activated. After repeated training and tuning, the average absolute error of the final preset neural network model on the test set is controlled within ±3 mmHg, meeting the requirements of the medical device standard.

[0071] To construct a high-precision blood pressure prediction model, the neural network structure is systematically trained. The specific implementation process is as follows:

[0072] First, construct the basic architecture of the neural network. The input layer contains 22 neurons corresponding to 22-dimensional input features: frequency-domain feature values (main frequency, frequency band width, spectral centroid, proportion of band energy, spectral peak ratio), amplitude feature values (maximum signal amplitude, average amplitude, amplitude standard deviation, waveform factor, peak factor, pulse factor), time-domain feature values (signal duration, rise time, fall time, peak occurrence time, zero-crossing rate, first-order difference), as well as the initial systolic blood pressure and the initial diastolic blood pressure.

[0073] The first hidden layer is configured with 64 neurons. For the hidden layer: ReLU activation function, f(x) = max(0, x), where x is the neuron input value, and the Dropout technique (dropout rate 0.3) is applied. The second hidden layer is configured with 32 neurons, also using the ReLU activation function and Dropout (dropout rate 0.3). For the output layer: linear activation function f(x) = x. The output layer contains 2 neurons corresponding to the target systolic blood pressure and the target diastolic blood pressure, respectively, and a linear activation function is used.

[0074] In the data preprocessing stage, the input features are standardized. The StandardScaler is used to transform each feature into a standard normal distribution with a mean of 0 and a standard deviation of 1: X_standardized = (X - μ) / σ, where X is the original feature value, μ is the feature mean, μ = (1 / n)Σxi, n is the number of samples, and σ is the feature standard deviation: The processed dataset is randomly divided into a training set, a validation set, and a test set in a ratio of 7:2:1.

[0075] The training process uses the mini-batch stochastic gradient descent algorithm, and the specific configuration is as follows:

[0076] The batch size is set to 64, that is, 64 samples are used to update the parameters each time.

[0077] The initial learning rate is set to 0.001, and the Adam optimizer is used (parameter configuration: beta1 = 0.9, beta2 = 0.999, epsilon = 1e-8).

[0078] The mean squared error is used as the loss function: MSE = (1 / n)Σ(y_pred - y_true)^2, where n is the number of samples, y_pred is the predicted blood pressure value by the model, and y_true is the standard blood pressure value.

[0079] L2 regularization term: L2_loss = λ * Σ(w^2), where: λ is the regularization coefficient, set to 0.01, w is the weight parameter of each layer of the network, and the total loss function: Total_loss = MSE + L2_loss.

[0080] The training iteration process includes:

[0081] Forward propagation: The input features are calculated through each layer of the network to obtain the predicted blood pressure value.

[0082] Backward propagation: Calculate the gradient of the loss function with respect to the parameters of each layer.

[0083] Parameter update: Use the Adam optimizer to update the network parameters according to the calculated gradients.

[0084] Validation and evaluation: Calculate the loss value on the validation set after each epoch.

[0085] Early stopping mechanism: When the validation set loss has not improved for 5 consecutive epochs, save the current optimal model parameters and stop training.

[0086] Final model evaluation criteria:

[0087] Mean absolute error (MAE): Systolic blood pressure error ≤ ±3 mmHg, diastolic blood pressure error ≤ ±2 mmHg.

[0088] Root Mean Square Error (RMSE): RMSE of systolic blood pressure ≤ 4 mmHg, RMSE of diastolic blood pressure ≤ 3 mmHg.

[0089] Correlation coefficient (R 2 ): R of systolic blood pressure and diastolic blood pressure 2 should both be greater than 0.95.

[0090] The preset neural network model after training should meet the following performance indicators:

[0091] Accuracy on the test set: The error of more than 95% of the measurement results is within the range of ±5 mmHg.

[0092] Model stability: The standard deviation of continuous measurements is less than 2 mmHg.

[0093] Generalization ability: It can maintain stable measurement accuracy for testers of different ages, genders, and physical conditions.

[0094] Suppose the measurement data of 1000 subjects are collected, and each subject is measured 3 times, obtaining a total of 3000 groups of sample data. The input features of each group of samples specifically include: Frequency domain feature values: Main frequency (35 Hz), Frequency band width (45 Hz), Spectrum centroid (42 Hz), Band energy ratio (0.65), Spectrum peak ratio (0.78); Amplitude feature values: Maximum signal amplitude (0.95), Average amplitude (0.45), Amplitude standard deviation (0.15), Waveform factor (1.25), Peak factor (2.1), Pulse factor (2.3); Time domain feature values: Signal duration (0.8 s), Rise time (0.3 s), Fall time (0.5 s), Peak occurrence time (0.25 s), Zero crossing rate (120 Hz), First order difference (0.05).

[0095] Initial blood pressure values: Initial systolic blood pressure (135 mmHg), Initial diastolic blood pressure (85 mmHg), and the corresponding output features are measured by a standard sphygmomanometer: Standard systolic blood pressure (130 mmHg) and Standard diastolic blood pressure (80 mmHg)

[0096] After data preprocessing, the 3000 groups of samples are divided according to the ratio of 7:2:1: Training set: 2100 groups, Validation set: 600 groups, Test set: 300 groups.

[0097] Record of the model training process: First round of training (Epoch1): Training set loss: MSE = 25.6;

[0098] Validation set loss: MSE = 23.8; MAE of systolic blood pressure = ±5.2 mmHg; MAE of diastolic blood pressure = ±4.1 mmHg.

[0099] Mid-training (Epoch 50): Training set loss: MSE = 12.3, Validation set loss: MSE = 11.9, Systolic blood pressure MAE = ±3.8 mmHg, Diastolic blood pressure MAE = ±2.9 mmHg.

[0100] Training completed (Epoch 100): Training set loss: MSE = 8.2, Validation set loss: MSE = 8.5, Systolic blood pressure MAE = ±2.8 mmHg, Diastolic blood pressure MAE = ±1.9 mmHg.

[0101] Final performance evaluation on the test set: Systolic blood pressure measurement results: Mean absolute error: ±2.8 mmHg, Root mean square error: 3.2 mmHg, Correlation coefficient R 2 : 0.96.

[0102] Diastolic blood pressure measurement results: Mean absolute error: ±1.9 mmHg, Root mean square error: 2.3 mmHg, Correlation coefficient R 2 : 0.97.

[0103] Test case verification: Input new measurement data: Frequency domain eigenvalues: Main frequency (38 Hz), Frequency band width (42 Hz), Spectrum centroid (40 Hz), Band energy ratio (0.68), Spectrum peak ratio (0.75); Amplitude eigenvalues: Signal maximum amplitude (0.92), Average amplitude (0.48), Amplitude standard deviation (0.14), Shape factor (1.28), Crest factor (2.0), Pulse factor (2.2); Time domain eigenvalues: Signal duration (0.75 s), Rise time (0.28 s), Fall time (0.47 s), Peak occurrence time (0.23 s), Zero crossing rate (125 Hz), First order difference (0.04).

[0104] Initial blood pressure values: Initial systolic blood pressure (140 mmHg), Initial diastolic blood pressure (88 mmHg).

[0105] Model prediction results: Predicted systolic blood pressure: 133 mmHg (Actual value: 132 mmHg, Error: +1 mmHg); Predicted diastolic blood pressure: 82 mmHg (Actual value: 81 mmHg, Error: +1 mmHg).

[0106] After obtaining the preset neural network model, it further includes: obtaining multiple groups of verification sample data, where each group of verification sample data includes verification feature parameters of multiple Korotkoff sound signals, verification initial systolic blood pressure, verification initial diastolic blood pressure, as well as verification standard systolic blood pressure and verification standard diastolic blood pressure measured by a standard blood pressure measurement device; inputting the verification feature parameters, verification initial systolic blood pressure, and verification initial diastolic blood pressure in each group of verification sample data into the preset neural network model to obtain predicted systolic blood pressure and predicted diastolic blood pressure; calculating a first difference between the predicted systolic blood pressure and the verification standard systolic blood pressure, and a second difference between the predicted diastolic blood pressure and the verification standard diastolic blood pressure; when both the first difference and the second difference meet the preset requirements, it is determined that the preset neural network model meets the usage requirements. If there is a situation where the first difference and / or the second difference do not meet the preset requirements, the neural network model is retrained until a preset neural network model that meets the usage requirements is obtained.

[0107] In one example, an independent verification sample data set is collected, and these data come from subjects who did not participate in model training to ensure the objectivity of verification. For each group of verification samples, the verification feature parameters of Korotkoff sound signals (including frequency domain eigenvalue, amplitude eigenvalue, and time domain eigenvalue) are extracted using the same method as the training data, the verification initial systolic blood pressure and verification initial diastolic blood pressure are recorded, and at the same time, a medical mercury sphygmomanometer is used to measure the verification standard systolic blood pressure and verification standard diastolic blood pressure as reference values. The verification feature parameters, verification initial systolic blood pressure, and verification initial diastolic blood pressure are input into the trained preset neural network model to obtain predicted systolic blood pressure and predicted diastolic blood pressure. According to the requirements of medical device standards, the difference between the predicted value and the standard value is calculated: the first difference is the absolute error between the predicted systolic blood pressure and the verification standard systolic blood pressure, and the second difference is the absolute error between the predicted diastolic blood pressure and the verification standard diastolic blood pressure. The preset requirements are set as follows: in more than 95% of the measurement results, the first difference does not exceed ±8 mmHg, the second difference does not exceed ±5 mmHg, and the average errors do not exceed ±5 mmHg and ±3 mmHg respectively. When the measurement results of all verification samples meet these preset requirements, it can be confirmed that the preset neural network model meets the usage requirements; if there is a situation that does not meet the preset requirements, the error cause needs to be analyzed. Possible optimization measures include: increasing the number of training samples, adjusting network structure parameters, optimizing the feature extraction method, or modifying the training strategy, etc., and then retraining the model until it meets the usage requirements.

[0108] S105, input multiple feature parameters, initial systolic blood pressure, and initial diastolic blood pressure into the preset neural network model to obtain the target systolic blood pressure and target diastolic blood pressure of the person to be measured, and use the target systolic blood pressure and target diastolic blood pressure as the blood pressure measurement results of the person to be measured.

[0109] Specifically, in order to improve the accuracy of blood pressure measurement, the preset neural network model is used to deeply analyze and process the feature parameters and initial blood pressure values obtained in the foregoing steps.

[0110] The preset neural network model adopts a multi-layer perceptron structure, including an input layer, two hidden layers and an output layer. The number of neurons in the input layer corresponds to the dimension of the feature parameters. The two hidden layers contain 64 and 32 neurons respectively, and the output layer contains 2 neurons corresponding to the target systolic blood pressure and the target diastolic blood pressure. The model uses ReLU as the activation function, is trained by the backpropagation algorithm, and uses the mean square error as the loss function.

[0111] In practical applications, the extracted feature parameters (including frequency domain eigenvalues, amplitude eigenvalues and time domain eigenvalues), together with the initial systolic blood pressure and the initial diastolic blood pressure, are normalized and then input into the neural network model. The model will calculate more accurate blood pressure values based on the complex non-linear relationships of these input parameters. The neural network model uses a large amount of clinically verified data during the training stage, including patient data of different ages, genders and physical conditions, to ensure that the model has good generalization ability. The target systolic blood pressure and target diastolic blood pressure output by the model are used as the final blood pressure measurement results. This neural network-based blood pressure measurement method has higher accuracy compared with the traditional Korotkoff sound method, can better handle individual differences and the influence of the measurement environment, and at the same time maintains the stability and repeatability of the measurement process.

[0112] In this way, not only the accuracy of blood pressure measurement is improved, but also the reliability of the measurement results is enhanced, providing a more valuable reference basis for clinical diagnosis.

[0113] Based on the above embodiments, as an alternative embodiment, in S105, inputting multiple feature parameters, the initial systolic blood pressure and the initial diastolic blood pressure into the preset neural network model to obtain the target systolic blood pressure and the target diastolic blood pressure of the person to be measured specifically includes S601 - S603:

[0114] S601, through the input layer of the preset neural network model, input multiple feature parameters, the initial systolic blood pressure and the initial diastolic blood pressure into the preset neural network model.

[0115] S602, using the hidden layer of the preset neural network model, perform feature extraction and non-linear transformation on the input multiple feature parameters, the initial systolic blood pressure and the initial diastolic blood pressure.

[0116] S603, through the output layer of the preset neural network model, obtain the target systolic blood pressure and the target diastolic blood pressure of the person to be measured.

[0117] Specifically, the extracted 22-dimensional feature parameters (including frequency domain eigenvalues, amplitude eigenvalues, time domain eigenvalues) and the initial systolic blood pressure and the initial diastolic blood pressure are input into the input layer of the neural network, and each input neuron corresponds to a feature parameter. After the data enters the input layer, it is first normalized so that all feature data are distributed within the same numerical range.

[0118] Next, the data flows through two hidden layers for deep feature extraction and non-linear transformation. The 64 neurons in the first hidden layer process the input data through the ReLU activation function to extract high-dimensional feature representations; the 32 neurons in the second hidden layer further reduce the dimension and combine the features, and enhance the non-linear expression ability of the model through the ReLU activation function. During the processing of the hidden layer, the Dropout mechanism randomly masks some neurons (dropout rate 0.3) to prevent the model from overfitting, and at the same time, the L2 regularization technique controls the scale of the weight parameters. Finally, the feature data processed by the hidden layer is passed to the output layer, and the 2 neurons in the output layer calculate the target systolic blood pressure and target diastolic blood pressure respectively through the linear activation function.

[0119] Taking an actual case as an example, when the input features include 22 parameters such as the main frequency of 35Hz and the frequency band width of 45Hz, after the standardization of the input layer and the feature extraction and non-linear transformation of the hidden layer, the prediction results of the target systolic blood pressure of 132 mmHg and the target diastolic blood pressure of 81 mmHg are finally obtained in the output layer. This multi-layer processing method based on a deep neural network can make full use of the multi-dimensional feature information in the blood pressure measurement process, realize the accurate mapping from the feature space to the blood pressure value, and ensure the accuracy and reliability of blood pressure prediction.

[0120] After taking the target systolic blood pressure and target diastolic blood pressure as the blood pressure measurement results of the person to be measured, it further includes:

[0121] Compare the target systolic blood pressure and target diastolic blood pressure with the preset normal ranges of systolic blood pressure and diastolic blood pressure respectively; when the target systolic blood pressure exceeds the preset normal range of systolic blood pressure and / or the target diastolic blood pressure exceeds the preset normal range of diastolic blood pressure, generate a blood pressure abnormality prompt message; display the target systolic blood pressure, target diastolic blood pressure and blood pressure abnormality prompt message on the display screen.

[0122] In one example, first, according to the blood pressure classification standard of the World Health Organization (WHO), the normal range of systolic blood pressure is set to 90-140mmHg, and the normal range of diastolic blood pressure is set to 60-90mmHg. After the target systolic blood pressure and target diastolic blood pressure are calculated by the neural network model, the system automatically compares these measured values with the preset normal range. To illustrate with an actual case, suppose that the target systolic blood pressure is 155mmHg and the target diastolic blood pressure is 85mmHg in a certain measurement. Since the target systolic blood pressure of 155mmHg exceeds the preset upper limit of the normal range of 140mmHg, and the target diastolic blood pressure of 85mmHg is within the normal range, the system will generate a blood pressure abnormality prompt message of "Warning: Systolic blood pressure is high". At the same time, the measurement results are clearly displayed on the display: systolic blood pressure 155mmHg (marked in red), diastolic blood pressure 85mmHg, and blood pressure abnormality prompt information is displayed in conjunction. This real-time monitoring and prompt mechanism can help users understand their blood pressure status in a timely manner.

[0123] Based on the above method, the present application also discloses an intelligent blood pressure measurement system based on the electronic Korotkoff sound method, such as Figure 2 As shown, Figure 2 The present invention is a schematic diagram of the structure of an intelligent blood pressure measurement system based on the electronic Korotkoff sound method provided in an embodiment of the present invention, wherein the device comprises: a control module, an acquisition module, a determination module, an extraction module and an output module; wherein the control module is used to increase the air pressure in the cuff wrapped around the upper arm of the person to be measured to a preset pressure and then deflate the cuff; the acquisition module is used to acquire multiple Korotkoff sound signals from the brachial artery and the air pressure in the cuff corresponding to each Korotkoff sound signal during the process of deflation of the cuff; the determination module is used to obtain the first occurrence of a plurality of Korotkoff sound signals according to the pressure in the cuff; the output ... The first air pressure in the cuff when a Korotkoff sound signal appears is used to determine the initial systolic pressure of the person to be tested, and the initial diastolic pressure of the person to be tested is determined according to the second air pressure in the cuff when the last Korotkoff sound signal appears among multiple Korotkoff sound signals; an extraction module is used to extract feature values of multiple Korotkoff sound signals to obtain multiple feature parameters; an output module is used to input the multiple feature parameters, the initial systolic pressure and the initial diastolic pressure into a preset neural network model to obtain the target systolic pressure and target diastolic pressure of the person to be tested, and use the target systolic pressure and target diastolic pressure as the blood pressure measurement results of the person to be tested.

[0124] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0125] Please refer to Figure 3 , which provides a schematic structural diagram of an electronic device for an embodiment of the present application. As Figure 3 shown, the electronic device 1000 may include: at least one processor 1001, at least one network interface 1004, a user interface 1003, a memory 1005, and at least one communication bus 1002.

[0126] Among them, the communication bus 1002 is used to implement connection communication between these components.

[0127] Among them, the user interface 1003 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 1003 may further include a standard wired interface and a wireless interface.

[0128] Among them, the network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0129] Among them, the processor 1001 may include one or more processing cores. The processor 1001 connects various parts within the entire server using various interfaces and lines, and by running or executing instructions, programs, code sets, or instruction sets stored in the memory 1005, as well as calling data stored in the memory 1005, it performs various functions of the server and processes data. Optionally, the processor 1001 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The processor 1001 may integrate a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. in a combination of one or several. Among them, the CPU mainly processes the operating system, user interface, and application programs, etc.; the GPU is responsible for rendering and drawing the content required to be displayed on the display screen; the modem is used to process wireless communication. It can be understood that the above-mentioned modem may not be integrated into the processor 1001 and may be implemented separately by a single chip.

[0130] Among them, the memory 1005 may include a Random Access Memory (RAM), or may also include a Read-Only Memory. Optionally, the memory 1005 includes a non-transitory computer-readable storage medium. The memory 1005 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1005 may include a program storage area and a data storage area. Among them, the program storage area can store instructions for implementing the operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area can store the data involved in the above-mentioned method embodiments. Optionally, the memory 1005 may also be at least one storage device located far from the aforementioned processor 1001. As Figure 3 shown, in the memory 1005 as a computer storage medium, it may include an operating system, a network communication module, a user interface module, and an application program for an intelligent blood pressure measurement method based on the electronic Korotkoff sound method.

[0131] In Figure 3 the electronic device 1000 shown, the user interface 1003 is mainly used to provide an input interface for the user to obtain the data input by the user; and the processor 1001 can be used to call the application program for an intelligent blood pressure measurement method based on the electronic Korotkoff sound method stored in the memory 1005. When executed by one or more processors, the electronic device is caused to execute the method as described in one or more of the above embodiments.

[0132] An electronic device-readable storage medium stores instructions. When executed by one or more processors, the electronic device is caused to execute the method as described in one or more of the above embodiments.

[0133] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present application is not limited by the described action sequence, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0134] In the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0135] In several embodiments provided in this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling, direct coupling, or communication connection to each other can be through some service interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical or other form.

[0136] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0137] In addition, in each embodiment of this application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0138] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of this application. The aforementioned memory includes: various media such as USB flash drives, mobile hard disks, magnetic disks, or optical discs that can store program codes.

[0139] The above are only exemplary embodiments of the present disclosure and should not be used to limit the scope of the present disclosure. That is, all equivalent changes and modifications made in accordance with the teachings of the present disclosure still fall within the scope covered by the present disclosure. Those skilled in the art will easily think of other implementation schemes of the present disclosure after considering the specification and practicing the present disclosure. This application aims to cover any variations, uses, or adaptation changes of the present disclosure, and these variations, uses, or adaptation changes follow the general principles of the present disclosure and include common general knowledge or conventional technical means in the technical field not recorded in the present disclosure. The specification and embodiments are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. An intelligent blood pressure measurement method based on the electronic Korotkoff sound method, characterized in that The method includes: increasing the air pressure in the cuff wrapped around the upper arm of the person to be measured to a preset pressure, and then deflating the cuff; during the deflation of the cuff, obtaining a plurality of Korotkoff sound signals from the brachial artery and the air pressure in the cuff corresponding to each of the Korotkoff sound signals; determining the initial systolic blood pressure of the person to be measured according to the first air pressure in the cuff when the Korotkoff sound signal first appears in the plurality of Korotkoff sound signals, and determining the initial diastolic blood pressure of the person to be measured according to the second air pressure in the cuff when the Korotkoff sound signal appears last in the plurality of Korotkoff sound signals; extracting eigenvalue of the plurality of Korotkoff sound signals to obtain a plurality of characteristic parameters, including: performing Fourier transform on each of the Korotkoff sound signals to obtain the frequency domain eigenvalue of each of the Korotkoff sound signals; performing amplitude normalization processing on each of the Korotkoff sound signals to obtain the amplitude eigenvalue of each of the Korotkoff sound signals; obtaining the time domain eigenvalue of each of the Korotkoff sound signals according to the duration of each of the Korotkoff sound signals; taking the frequency domain eigenvalue, the amplitude eigenvalue and the time domain eigenvalue as the plurality of characteristic parameters of each of the Korotkoff sound signals. Obtain multiple groups of sample data, each group of the sample data including the characteristic parameters of a plurality of Korotkoff sound signals, the initial systolic blood pressure, the initial diastolic blood pressure, the standard systolic blood pressure and the standard diastolic blood pressure measured by a standard blood pressure measuring device; taking the characteristic parameters, the initial systolic blood pressure and the initial diastolic blood pressure in each group of the sample data as input features, and taking the standard systolic blood pressure and the standard diastolic blood pressure as output features; training a neural network structure according to the input features and the output features to obtain a preset neural network model. Input the multiple characteristic parameters, the initial systolic blood pressure and the initial diastolic blood pressure into the preset neural network model to obtain the target systolic blood pressure and the target diastolic blood pressure of the person to be measured, including: inputting the multiple characteristic parameters, the initial systolic blood pressure and the initial diastolic blood pressure into the preset neural network model through the input layer of the preset neural network model; using the hidden layer of the preset neural network model to perform feature extraction and non-linear transformation on the input multiple characteristic parameters, the initial systolic blood pressure and the initial diastolic blood pressure; obtaining the target systolic blood pressure and the target diastolic blood pressure of the person to be measured through the output layer of the preset neural network model; taking the target systolic blood pressure and the target diastolic blood pressure as the blood pressure measurement result of the person to be measured.

2. The intelligent blood pressure measurement method based on the electronic Korotkoff sound method according to claim 1, wherein The obtaining of the plurality of Korotkoff sound signals from the brachial artery and the air pressure in the cuff corresponding to each of the Korotkoff sound signals includes: during the deflation of the cuff, receiving the plurality of Korotkoff sound signals generated by the pulse beating of the brachial artery collected by a sound sensor arranged in the cuff, and receiving the air pressure in the cuff corresponding to each of the Korotkoff sound signals sent by a pressure sensor arranged in the cuff.

3. The intelligent blood pressure measurement method based on the electronic Korotkoff sound method according to claim 1, characterized in that, After obtaining the preset neural network model, the method further includes: obtaining multiple groups of verification sample data, where each group of the verification sample data includes verification feature parameters of multiple Korotkoff sound signals, an initial verification systolic blood pressure, an initial verification diastolic blood pressure, and a verification standard systolic blood pressure and a verification standard diastolic blood pressure measured by a standard blood pressure measurement device; inputting the verification feature parameters, the initial verification systolic blood pressure, and the initial verification diastolic blood pressure in each group of the verification sample data into the preset neural network model to obtain a predicted systolic blood pressure and a predicted diastolic blood pressure; calculating a first difference between the predicted systolic blood pressure and the verification standard systolic blood pressure, and a second difference between the predicted diastolic blood pressure and the verification standard diastolic blood pressure; when both the first difference and the second difference meet the preset requirements, determining that the preset neural network model meets the usage requirements, and if there is a situation where the first difference and / or the second difference do not meet the preset requirements, retraining the neural network model until a preset neural network model that meets the usage requirements is obtained.

4. The intelligent blood pressure measurement method based on the electronic Korotkoff sound method according to claim 1, wherein After using the target systolic blood pressure and the target diastolic blood pressure as the blood pressure measurement results of the person to be measured, the method further includes: comparing the target systolic blood pressure and the target diastolic blood pressure with a preset normal range of systolic blood pressure and a preset normal range of diastolic blood pressure respectively; when the target systolic blood pressure exceeds the preset normal range of systolic blood pressure and / or the target diastolic blood pressure exceeds the preset normal range of diastolic blood pressure, generating a blood pressure abnormality prompt message; displaying the target systolic blood pressure, the target diastolic blood pressure, and the blood pressure abnormality prompt message on a display screen.

5. An intelligent blood pressure measurement system based on the electronic Korotkoff sound method, characterized in that, The system includes: a control module, an acquisition module, a determination module, an extraction module, and an output module; wherein, the control module is configured to increase the air pressure in the cuff wrapped around the upper arm of the person to be measured to a preset pressure and then deflate the cuff; the acquisition module is configured to acquire a plurality of Korotkoff sound signals from the brachial artery and the air pressure in the cuff corresponding to each of the Korotkoff sound signals during the deflation of the cuff; the determination module is configured to determine the initial systolic blood pressure of the person to be measured according to the first air pressure in the cuff when the Korotkoff sound signal first appears among the plurality of Korotkoff sound signals, and determine the initial diastolic blood pressure of the person to be measured according to the second air pressure in the cuff when the Korotkoff sound signal appears last among the plurality of Korotkoff sound signals; the extraction module is configured to extract eigenvalue of the plurality of Korotkoff sound signals to obtain a plurality of characteristic parameters, including: performing Fourier transform on each of the Korotkoff sound signals to obtain the frequency domain eigenvalue of each of the Korotkoff sound signals; performing amplitude normalization processing on each of the Korotkoff sound signals to obtain the amplitude eigenvalue of each of the Korotkoff sound signals; obtaining the time domain eigenvalue of each of the Korotkoff sound signals according to the duration of each of the Korotkoff sound signals; using the frequency domain eigenvalue, the amplitude eigenvalue, and the time domain eigenvalue as the plurality of characteristic parameters of each of the Korotkoff sound signals; the output module is configured to obtain multiple sets of sample data, each set of the sample data including the characteristic parameters of a plurality of Korotkoff sound signals, the initial systolic blood pressure, the initial diastolic blood pressure, and the standard systolic blood pressure and the standard diastolic blood pressure measured by a standard blood pressure measuring device; using the characteristic parameters, the initial systolic blood pressure, and the initial diastolic blood pressure in each set of the sample data as input features, and using the standard systolic blood pressure and the standard diastolic blood pressure as output features; training a neural network structure according to the input features and the output features to obtain a preset neural network model; inputting the plurality of characteristic parameters, the initial systolic blood pressure, and the initial diastolic blood pressure into the preset neural network model to obtain the target systolic blood pressure and the target diastolic blood pressure of the person to be measured, including: inputting the plurality of characteristic parameters, the initial systolic blood pressure, and the initial diastolic blood pressure into the preset neural network model through the input layer of the preset neural network model; using the hidden layer of the preset neural network model to perform feature extraction and non-linear transformation on the input plurality of characteristic parameters, the initial systolic blood pressure, and the initial diastolic blood pressure; obtaining the target systolic blood pressure and the target diastolic blood pressure of the person to be measured through the output layer of the preset neural network model; using the target systolic blood pressure and the target diastolic blood pressure as the blood pressure measurement result of the person to be measured.

6. An electronic device, characterized in that, It includes a processor, a memory, a user interface, and a network interface. The memory is used to store instructions. The user interface and the network interface are used to communicate with other devices. The processor is used to execute the instructions stored in the memory so that the electronic device executes the method according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, A computer program is stored that can be loaded and executed by a processor to execute the method according to any one of claims 1-4.

Images