Blood pressure measurement method and apparatus, computer device, and storage medium

Abstract

A blood pressure measurement method and apparatus, a computer device, and a storage medium. The blood pressure measurement method comprises: acquiring blood pressure measurement information to form audio data (S110); preprocessing the audio data to obtain a preprocessing result (S120); inputting the preprocessing result into an automatic blood pressure generation model to generate a blood pressure value, so as to obtain a generation result (S130); and outputting the generation result (S140), wherein the automatic blood pressure generation model is obtained by training a deep learning model using blood pressure measurement information collected from a plurality of persons as a sample set.

Description

Blood pressure measurement methods, devices, computer equipment, and storage media

[0001] This application is based on and claims priority to Chinese Patent Application No. 202510042618.2, filed on January 10, 2025, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to blood pressure measurement technology, and more specifically to blood pressure measurement methods, devices, computer equipment, and storage media. Background Technology

[0003] With the rapid development of medical technology, electronic blood pressure monitors have gained widespread recognition and application in the market due to their convenience and intelligent features. The demand for electronic blood pressure monitors continues to grow, especially in areas such as home health management and routine monitoring in medical institutions. Among these, electronic blood pressure monitors based on the auscultation method (i.e., the Korotkoff sound method), which determine blood pressure values ​​by detecting the sound of arterial blood flow (Korotkoff sounds), have become one of the mainstream technologies.

[0004] The working principle of existing Korotkoff sound electronic blood pressure monitors is as follows: A cuff is wrapped around the patient's arm and pressurized by an air pump until the cuff pressure exceeds the maximum arterial pressure (systolic pressure). At this point, the artery is completely closed, blood cannot flow, and no sound is heard. Subsequently, the pressure in the cuff is slowly released, and the artery begins to partially open, producing Korotkoff sounds due to blood turbulence. The cuff pressure corresponding to the first appearance of a Korotkoff sound is the systolic pressure; and the cuff pressure corresponding to the complete disappearance of the Korotkoff sound is the diastolic pressure.

[0005] However, despite the important position of Korotkoff sound electronic blood pressure monitors in the field of blood pressure measurement, they still face many challenges and limitations:

[0006] Algorithms and Signal Processing: Currently, most Korotkoff sound electronic blood pressure monitors rely on processors for digital signal processing to analyze audio data. However, the choice of algorithm, the accuracy of feature extraction, and real-time processing capabilities directly affect the accuracy of blood pressure measurement. Imperfections in the algorithm or errors in signal processing may cause measurement results to deviate from the actual value.

[0007] Piezoelectric signal capture: Some blood pressure monitors attempt to measure blood pressure by capturing the mechanical vibration of the brachial artery pulsation using piezoelectric sensors. However, this method is highly susceptible to external vibration interference, and the vibration signal itself is relatively weak, increasing the uncertainty and error of the measurement.

[0008] Electret microphone applications: To improve measurement accuracy, some blood pressure monitors use electret microphones to pick up Korotkoff tone signals. However, the introduction of electret microphones increases the complexity of circuit design, requiring additional amplifiers and filters for signal processing. Improper circuit design can easily introduce additional noise and distortion, affecting the accuracy of measurement results. Furthermore, the parameters of electret microphones (such as sensitivity and frequency response) are inconsistent in mass production, further increasing measurement errors.

[0009] Environmental factors and individual differences: Blood pressure measurements are also affected by environmental noise and individual patient differences (such as age, gender, body type, degree of arteriosclerosis, etc.). These factors may cause changes in the intensity, frequency, and duration of Korotkoff sounds, thus affecting the accuracy and reliability of the measurement.

[0010] Given the aforementioned limitations, the industry urgently needs to develop new, efficient, and accurate blood pressure measurement technologies to improve the accuracy of blood pressure measurement and user experience. Application content

[0011] The purpose of this application is to overcome the shortcomings of the prior art and to provide a method, device, equipment and medium for measuring blood pressure.

[0012] To solve the above-mentioned technical problems, this application adopts the following technical solution:

[0013] Firstly, it provides methods for measuring blood pressure, including:

[0014] Obtain blood pressure measurement information to generate audio data;

[0015] The audio data is preprocessed to obtain the preprocessed result;

[0016] The preprocessing results are input into the automatic blood pressure generation model to generate blood pressure, thus obtaining the generated result;

[0017] Output the generated results;

[0018] The automatic blood pressure generation model is obtained by training a deep learning model using blood pressure measurement information from multiple people as a sample set.

[0019] Secondly, a blood pressure measuring device is provided, including:

[0020] An acquisition forming unit is used to acquire blood pressure measurement information to form audio data;

[0021] The preprocessing unit is used to preprocess the audio data to obtain the preprocessed result;

[0022] The automatic generation unit is used to input the preprocessed results into the automatic blood pressure generation model to generate blood pressure and obtain the generated result.

[0023] The output unit is used to output the generated results;

[0024] The automatic blood pressure generation model is obtained by training a deep learning model using blood pressure measurement information from multiple people as a sample set.

[0025] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described blood pressure measurement method.

[0026] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the above-described blood pressure measurement method.

[0027] The aforementioned blood pressure measurement method ensures the integrity and authenticity of the original data by acquiring blood pressure measurement information and generating audio data. Preprocessing the audio data effectively removes noise and interference, improving data purity and usability. A deep learning model is then introduced as an automatic blood pressure generation model. This model is trained using blood pressure measurement information from multiple individuals as a sample set, learning rich blood pressure features to achieve high-precision prediction and generation of blood pressure values. The preprocessed data is directly input into the automatic blood pressure generation model to generate blood pressure values ​​without human intervention, realizing intelligent and automated blood pressure measurement. This not only improves measurement efficiency but also reduces errors caused by human operation, making blood pressure measurement more convenient and accurate.

[0028] The present application will be further described below with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0029] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 is a schematic flowchart of the blood pressure measurement method provided in the embodiment of this application;

[0031] Figure 2 is a schematic block diagram of the blood pressure measurement structure provided in an embodiment of this application;

[0032] Figure 3 is a schematic block diagram of the automatic blood pressure generation model provided in the embodiments of this application;

[0033] Figure 4 is a schematic block diagram of the blood pressure measuring device provided in the embodiments of this application;

[0034] Figure 5 is a schematic diagram of the structure of the computer device in an embodiment of this application. Detailed Implementation

[0035] 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.

[0036] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0037] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0038] It should also be further understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0039] Please refer to the specific embodiments shown in Figures 1 to 3. This application discloses a blood pressure measurement method, including the following steps:

[0040] S110, acquire blood pressure measurement information to generate audio data;

[0041] Specifically, the blood pressure measurement method is based on a blood pressure measurement structure, as shown in Figure 2. This structure includes an air pump, an air valve, a cuff, a sound sensor, a display screen, key input, an ADC, a pressure sensor, and a microcontroller. The cuff includes an inflatable air bladder and a stethoscope attached to the inner surface of the cuff. The inflatable air bladder is connected to the multi-way valve, air pump, air valve, and pressure sensor via a conduit. The stethoscope is connected to the sound sensor via another conduit. The microcontroller controls the entire blood pressure measurement process and external input / output signals, including air pump inflation, air valve deflation, display screen output, key input / output detection, receiving signals from the sound sensor, and reading signals from the pressure sensor.

[0042] More specifically, the sound sensor employs a MEMS sound sensor module, which integrates a MEMS microphone to capture sound signals (sound waves) from the surrounding environment. The sound is picked up by the microphone in the form of a sound pressure signal. The captured analog sound pressure signal is converted into a digital signal via an ADC (analog-to-digital converter), and the converted digital audio signal is output through an I²S interface for further processing by a microcontroller or other devices.

[0043] In one embodiment, the step of acquiring blood pressure measurement information to form audio data includes: using a microcontroller to read pressure data from a pressure sensor whose voltage signal is converted via an ADC to control inflation and deflation; at the start of deflation, a sound sensor connected to a stethoscope begins to record sound; and at the end of deflation, the recording ends to obtain blood pressure measurement information, which is then stored in the microcontroller to form audio data.

[0044] Specifically, after system startup, the microcontroller first initializes, including ADC calibration and sensor self-testing, to ensure measurement accuracy. The microcontroller controls the airbag to rapidly inflate to a preset pressure level while continuously reading data from the pressure sensor via the ADC to monitor pressure changes. Once the maximum inflation pressure is reached, the microcontroller controls the airbag to begin slow deflation. At this time, the sound sensor activates simultaneously, starting to record arterial pulsation sounds. During deflation, the microcontroller continuously monitors pressure data and determines systolic and diastolic blood pressure based on a preset algorithm (such as recognizing the first Korotkoff sound and the final disappearance sound). When the last Korotkoff sound is detected, the microcontroller instructs the sound sensor to stop recording. After recording, the microcontroller stores the recorded audio data (which may be compressed to reduce storage space) in internal or external storage.

[0045] More specifically, by combining data from pressure and sound sensors, systolic and diastolic blood pressure can be determined more accurately, reducing human error. The entire measurement process is controlled by a microcontroller, eliminating the need for manual operation of the stethoscope or data recording, thus improving the automation level and user convenience. The storage of audio data enables subsequent data analysis, such as further optimizing measurement results through software algorithms, or using it for remote medical consultation and diagnosis. It is not only applicable to traditional blood pressure monitors but can also be integrated into smart wearable devices to meet blood pressure monitoring needs in various scenarios.

[0046] S120, preprocess the audio data to obtain the preprocessed result;

[0047] In one embodiment, the preprocessing of the audio data to obtain a preprocessing result includes: decoding the audio data to obtain time-related data, performing Fourier transform on each frame of this data to obtain a frequency distribution, and finally combining this frequency distribution according to time to obtain a spectrum, i.e., the preprocessing result.

[0048] Specifically, first, the audio data to be processed is acquired from the microcontroller. If the audio data is in a compressed format (such as MP3, WAV, FLAC, etc.), it needs to be decompressed or decoded to convert it into uncompressed PCM (Pulse Code Modulation) format or similar raw audio data. The purpose of this step is to obtain audio signal data directly related to time, i.e., audio waveform data. Then, the decoded audio data is divided into multiple short time segments in chronological order, each time segment being called a frame. The purpose of framing is to decompose long-time audio signals into shorter, more easily processed signals. Based on framing, a window function (such as Hanning window, Hamming window, etc.) is applied to each frame of audio data to reduce spectral leakage at frame edges. A Fast Fourier Transform is performed on each windowed frame of audio data to convert the time-domain signal into a frequency-domain signal, obtaining the frequency distribution (i.e., spectrum) of that frame. The spectrum of each frame is arranged in chronological order to form a two-dimensional matrix, where the horizontal axis represents time, the vertical axis represents frequency, and the values ​​in the matrix represent the signal strength (i.e., spectral energy) at the corresponding time and frequency. Using graphical tools or algorithms, the above two-dimensional matrix is ​​converted into a visualized spectrum. A spectrogram can visually demonstrate the frequency characteristics of an audio signal as it changes over time.

[0049] More specifically, Fourier transform converts audio signals from the time domain to the frequency domain, effectively extracting frequency features, which is crucial for subsequent signal analysis and feature recognition. Furthermore, the choice of frame segmentation and windowing function balances time and frequency resolution, ensuring the spectrogram captures rapid changes in the audio signal while accurately reflecting its frequency components. Additionally, spectrogram generation makes audio signal feature analysis more intuitive and convenient, aiding in the discovery of hidden information, anomalous signals, or specific patterns. This technology is not only suitable for audio data preprocessing in blood pressure measurement but can also be widely applied in speech recognition, music analysis, noise monitoring, fault diagnosis, and other fields, demonstrating broad practical value.

[0050] S130, input the preprocessing results into the automatic blood pressure generation model to generate blood pressure, so as to obtain the generated result;

[0051] Specifically, referring to the automatic blood pressure generation model shown in Figure 3, a large sample set containing blood pressure measurements and corresponding audio data was collected during the development phase. Using this sample set, a blood pressure prediction model was trained using machine learning or deep learning algorithms (such as convolutional neural networks, CNNs). During training, the model learns the complex relationship between audio signals and blood pressure values. The trained model was then installed on target devices, such as smart bracelets, health monitors, and blood pressure measurement devices. When the device receives audio data from the user, it inputs it into the deployed model. The convolutional layers in the model extract features from the input audio data. Through sliding and convolution operations of multiple convolutional kernels, key features in the audio signal, such as frequency and amplitude, are extracted. The parameters in the model (such as weights and biases) have been optimized during training to accurately map the extracted features to blood pressure values. Through a series of nonlinear transformations and linear combinations, an accurate blood pressure prediction result is finally obtained. The model outputs the systolic and diastolic blood pressure values ​​and then displays the calculated blood pressure values ​​to the user in real time on the device's display screen.

[0052] More specifically, models trained using machine learning or deep learning algorithms can accurately capture the complex relationship between audio signals and blood pressure values, achieving high-precision blood pressure prediction. The entire analysis process is fast and efficient, providing blood pressure prediction results shortly after the user records audio, enhancing the user experience. Using audio data for blood pressure prediction avoids the invasiveness and discomfort of traditional blood pressure measurement methods, improving the convenience and comfort of measurement. This technology is not only suitable for home health monitoring but can also be applied to medical institutions, sports and fitness, and other fields, possessing broad market prospects and practical value.

[0053] S140, Output the generated result;

[0054] In one embodiment, the output generation result includes: displaying the generation result in a visual manner.

[0055] Specifically, blood pressure reports are generated based on the trained model and presented visually for easy viewing and comparison by users. The blood pressure reports will be generated using custom templates, and the presentation format can include tables, charts, and other visual representations.

[0056] The automatic blood pressure generation model is obtained by training a deep learning model using blood pressure measurement information from multiple people as a sample set.

[0057] In one embodiment, the automatic blood pressure generation model is obtained by training a deep learning model using blood pressure measurement information from multiple individuals as a sample set, including:

[0058] Blood pressure measurement information from multiple individuals was collected and labeled to obtain a sample set;

[0059] First, blood pressure measurements were collected from a varying number of volunteers. This data included, but was not limited to, systolic and diastolic blood pressure, along with the corresponding measurement time. Ensuring sample diversity, encompassing individuals of different ages, genders, and health conditions, improved the model's generalization ability. Next, the collected blood pressure measurements were labeled, marking the specific values ​​of systolic and diastolic blood pressure, as well as their corresponding time points. This labeled data constituted the sample set required to train the deep learning model.

[0060] Build deep learning models;

[0061] Choose a suitable deep learning framework for processing time series data, such as TensorFlow or PyTorch. Build a deep learning model; the initial model structure can be a multilayer perceptron (MLP) or convolutional neural network (CNN) as a foundation for preliminary feature extraction.

[0062] Tuning parameters of deep learning models;

[0063] This involves adjusting the hyperparameters of the deep learning model (such as learning rate, batch size, number of network layers, etc.) to optimize model performance.

[0064] The parameter-tuned deep learning model is trained using a sample set to obtain a preliminarily trained model.

[0065] The process involves training the adjusted model using a labeled sample set, updating the model weights through backpropagation, until the model's performance on the validation set no longer shows significant improvement, thus obtaining the initially trained model.

[0066] The performance of the initial training model was evaluated and optimized to obtain an automatic blood pressure generation model.

[0067] The performance of the initially trained model was evaluated using metrics such as cross-validation, accuracy, recall, and F1 score. Based on the evaluation results, the model structure or training strategy was optimized, such as by adding regularization terms, adjusting the number of network layers, or adjusting the activation function, until the model performance reached a satisfactory level, forming the final automatic blood pressure generation model. Blood pressure measurement data was converted into a spectrogram, and time features were added to the spectrogram to preserve time series information. The spectrogram with added time features was input into a neural network for encoding, extracting deep information (micro-information), especially features in the time series. A two-layer bidirectional long short-term memory network (Bi-LSTM) was constructed, which can progressively process the features of each time point in the time dimension and capture long-term dependencies in the time series. A decoder module was designed to decode the output of the two-layer bidirectional LSTM network into target values, i.e., the predicted values ​​of systolic and diastolic blood pressure. By setting target values ​​(standard values), the model can be trained according to these standards, continuously adjusting the model parameters and improving prediction accuracy. The output of a two-layer bidirectional LSTM network is mapped to a target prediction space, which contains the prediction start and end times of systolic and diastolic blood pressure. Based on the prediction start and end times, the specific values ​​of systolic and diastolic blood pressure are determined, thereby realizing the automatic generation and prediction of blood pressure.

[0068] More specifically, through the training and optimization of deep learning models, individual systolic and diastolic blood pressure can be predicted more accurately, improving the precision of blood pressure measurement. The model can automatically process input blood pressure measurement data, completing blood pressure prediction and generation without manual intervention, thus improving work efficiency. Trained by collecting blood pressure measurement information from multiple individuals, the model has strong generalization ability and can be applied to blood pressure prediction for different populations. Furthermore, by incorporating temporal features and utilizing a two-layer bidirectional LSTM network, the model can better capture long-term dependencies in time series data, improving the stability of predictions.

[0069] The aforementioned blood pressure measurement method ensures the integrity and authenticity of the original data by acquiring blood pressure measurement information and generating audio data. Preprocessing the audio data effectively removes noise and interference, improving data purity and usability. A deep learning model is then introduced as an automatic blood pressure generation model. This model is trained using blood pressure measurement information from multiple individuals as a sample set, learning rich blood pressure features to achieve high-precision prediction and generation of blood pressure values. The preprocessed data is directly input into the automatic blood pressure generation model to generate blood pressure values ​​without human intervention, realizing intelligent and automated blood pressure measurement. This not only improves measurement efficiency but also reduces errors caused by human operation, making blood pressure measurement more convenient and accurate. Furthermore, due to the powerful learning and generalization capabilities of deep learning models, they can adapt to the blood pressure characteristics of different populations, including individual differences in age, gender, and body type, resulting in wider applicability and higher accuracy in practical applications. Additionally, the method in this application is based on existing electronic blood pressure monitor technology and deep learning technology, requiring no additional hardware investment; only software upgrades or development are needed. This reduces costs and makes the method easier to promote and apply in medical institutions, home health monitoring, and other fields.

[0070] Figure 4 is a schematic block diagram of a blood pressure measuring device 300 provided in an embodiment of this application. As shown in Figure 4, corresponding to the above-described blood pressure measuring method, this application also provides a blood pressure measuring device 300. The blood pressure measuring device 300 includes a unit for performing the above-described blood pressure measuring method, and the device can be configured in a server. Specifically, referring to Figure 4, the blood pressure measuring device 300 includes an acquisition and formation unit 301, a preprocessing unit 302, an automatic generation unit 303, and an output unit 304.

[0071] The acquisition forming unit 301 is used to acquire blood pressure measurement information to form audio data;

[0072] The preprocessing unit 302 is used to preprocess the audio data to obtain the preprocessing result;

[0073] The automatic generation unit 303 is used to input the preprocessing results into the automatic blood pressure generation model to generate blood pressure and obtain the generated result;

[0074] Output unit 304 is used to output the generated results.

[0075] In one embodiment, the acquisition and formation unit 301 controls the inflation and deflation by reading the pressure data of the voltage signal of the pressure sensor through the ADC conversion using a microcontroller. When deflation begins, the sound sensor connected to the stethoscope begins to record sound and ends the recording when deflation ends, so as to obtain blood pressure measurement information. Then, the formed audio data is stored in the microcontroller.

[0076] In one embodiment, the preprocessing unit 302 decodes the audio data to obtain time-related data, performs Fourier transform on each frame of this data to obtain a frequency distribution, and finally combines this frequency distribution according to time to obtain a spectrum, which is the preprocessing result.

[0077] In one embodiment, the blood pressure measuring device 300 further includes a model generation unit for training a deep learning model by collecting blood pressure measurement information from multiple people as a sample set.

[0078] In one embodiment, the model generation unit described above includes:

[0079] The sample set generation sub-unit is used to collect blood pressure measurement information from multiple people and label the blood pressure measurement information to obtain a sample set;

[0080] The model building subunit is used to build deep learning models;

[0081] The parameter tuning subunit is used to adjust the parameters of the deep learning model;

[0082] The training subunit is used to train the hyperparameter-tuned deep learning model using a sample set to obtain a pre-trained model.

[0083] The optimization sub-unit is used to evaluate and optimize the performance of the initial trained model to obtain an automatic blood pressure generation model.

[0084] In one embodiment, the output unit 304 is used to display the generated results in a visual manner.

[0085] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above-mentioned blood pressure measuring device 300 and each unit can be referred to the corresponding description in the foregoing method embodiments. For the sake of convenience and brevity, it will not be repeated here.

[0086] Please refer to Figure 5, which is a schematic block diagram of a computer device provided in an embodiment of this application. The computer device 500 is a server, wherein the server can be a standalone server or a server cluster composed of multiple servers.

[0087] Referring to Figure 5, the computer device 500 includes a processor 502, a memory, and a network interface 505 connected via a system bus 501. The memory may include a non-volatile storage medium 503 and internal memory 504.

[0088] The non-volatile storage medium 503 may store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions that, when executed, cause the processor 502 to perform a blood pressure measurement method.

[0089] The processor 502 provides computing and control capabilities to support the operation of the entire computer device 500.

[0090] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can perform a blood pressure measurement method.

[0091] The network interface 505 is used for network communication with other devices. Those skilled in the art will understand that the structure shown in Figure 5 is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device 500 to which the present application is applied. A specific computer device 500 may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0092] The processor 502 is used to run a computer program 5032 stored in the memory to perform the following steps:

[0093] The process involves acquiring blood pressure measurement information to generate audio data; preprocessing the audio data to obtain preprocessing results; inputting the preprocessing results into an automatic blood pressure generation model to generate blood pressure data; and outputting the generated results. The automatic blood pressure generation model is obtained by training a deep learning model using blood pressure measurement information from multiple individuals as a sample set.

[0094] It should be understood that in the embodiments of this application, the processor 502 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0095] It will be understood by those skilled in the art that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program includes program instructions and can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiments of the above methods.

[0096] Therefore, this application also provides a storage medium. This storage medium can be a computer-readable storage medium. The storage medium stores a computer program, wherein when executed by a processor, the computer program causes the processor to perform the following steps:

[0097] The process involves acquiring blood pressure measurement information to generate audio data; preprocessing the audio data to obtain preprocessing results; inputting the preprocessing results into an automatic blood pressure generation model to generate blood pressure data; and outputting the generated results. The automatic blood pressure generation model is obtained by training a deep learning model using blood pressure measurement information from multiple individuals as a sample set.

[0098] The storage medium can be any computer-readable storage medium capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), magnetic disk, or optical disk.

[0099] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0100] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of each unit is merely a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0101] The steps in the methods of this application embodiment can be adjusted, merged, or deleted according to actual needs. The units in the apparatus of this application embodiment can be merged, divided, or deleted according to actual needs. Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0102] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a storage medium. 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 the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0103] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for measuring blood pressure, characterized in that, include: Obtain blood pressure measurement information to generate audio data; The audio data is preprocessed to obtain the preprocessed result; The preprocessing results are input into the automatic blood pressure generation model to generate blood pressure, thus obtaining the generated result; Output the generated results; The automatic blood pressure generation model is obtained by training a deep learning model using blood pressure measurement information from multiple people as a sample set.

2. The blood pressure measurement method according to claim 1, characterized in that, The process of acquiring blood pressure measurement information to generate audio data includes: The microcontroller reads the pressure data from the voltage signal of the pressure sensor converted by the ADC to control the inflation and deflation. When deflation begins, the sound sensor connected to the stethoscope starts recording sound and stops recording when deflation ends, so as to obtain blood pressure measurement information. The audio data is then stored in the microcontroller.

3. The blood pressure measurement method according to claim 1, characterized in that, The preprocessing of audio data to obtain preprocessing results includes: The audio data is decoded to obtain time-related data. Then, each frame of this data is subjected to Fourier transform to obtain the frequency distribution. Finally, this frequency distribution is combined according to time to obtain a spectrum, which is the preprocessing result.

4. The blood pressure measurement method according to claim 1, characterized in that, The automatic blood pressure generation model is obtained by training a deep learning model using blood pressure measurement information from multiple individuals as a sample set, including: Blood pressure measurement information from multiple individuals was collected and labeled to obtain a sample set; Build deep learning models; Tuning parameters of deep learning models; The parameter-tuned deep learning model is trained using a sample set to obtain a preliminarily trained model. The performance of the initial training model was evaluated and optimized to obtain an automatic blood pressure generation model.

5. The blood pressure measurement method according to claim 1, characterized in that, The output generation results include: The generated results are displayed in a visual manner.

6. A blood pressure measuring device, characterized in that, include: An acquisition forming unit is used to acquire blood pressure measurement information to form audio data; The preprocessing unit is used to preprocess the audio data to obtain the preprocessed result; The automatic generation unit is used to input the preprocessed results into the automatic blood pressure generation model to generate blood pressure and obtain the generated result. The output unit is used to output the generated results; The automatic blood pressure generation model is obtained by training a deep learning model using blood pressure measurement information from multiple people as a sample set.

7. The blood pressure measuring device according to claim 6, characterized in that, In the acquisition and formation unit, a microcontroller reads the pressure data from the voltage signal of the pressure sensor converted by the ADC to control the inflation and deflation. When deflation begins, a sound sensor connected to the stethoscope begins to record sound and ends recording when deflation ends, so as to obtain blood pressure measurement information. Then, the audio data is stored in the microcontroller.

8. The blood pressure measuring device according to claim 6, characterized in that, In the preprocessing unit, the audio data is decoded to obtain time-related data, and then each frame of this data is subjected to Fourier transform to obtain the frequency distribution. Finally, this frequency distribution is combined according to time to obtain a spectrum, which is the preprocessing result.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the blood pressure measurement method as described in any one of claims 1 to 5.

10. A storage medium, wherein the computer-readable storage medium stores a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the blood pressure measurement method as described in any one of claims 1 to 5.

Images