Continuous blood pressure measurement method and device based on fingertip pulse waves

Abstract

The invention discloses a continuous blood pressure measurement method and device based on fingertip pulse waves. The device comprises a fingertip pulse wave signal acquisition module, a blood pressure modeling module, an embedded blood pressure calculation module and a complete machine system design module. The method comprises the following steps: a pulse wave signal acquisition module acquires a human fingertip pulse wave signal; the blood pressure modeling module is used for selecting and optimizing pulse wave characteristics of pulse wave time domain and frequency domain of the collected human fingertip pulse wave signals, and optimizing and training a support vector machine model for predicting blood pressure based on pulse waves of a public database and an experimental database; the embedded blood pressure calculation module completes deployment of a trained support vector machine model on a single-chip microcomputer, writes an input feature calculation function and a model output function, and completes prediction output of blood pressure and calculation of pulse rate and blood oxygen physiological parameters; the whole machine system design module carries out system program design, man-machine interaction key management, user and parameter display management and Bluetooth remote transmission of measurement data on the measurement device.

Description

Technical Field

[0001] The present invention belongs to the technical field of health monitoring, and in particular relates to a continuous blood pressure measurement method and device based on fingertip pulse waves. Background Art

[0002] Blood pressure (BP) refers to the lateral pressure per unit area of ​​the blood vessel wall when blood flows in the blood vessels. As a physiological indicator of the human body, it plays an important role in the diagnosis of cardiovascular diseases. Currently, commonly used non-invasive blood pressure measurement methods such as Korotkoff sound method and oscillometric method require the use of an inflated cuff to compress the upper arm or wrist, measure the pressure change during the cuff deflation process, and then calculate the single blood pressure measurement value through special equipment. Multiple measurements require an interval of 2-5 minutes and will cause arm discomfort. For clinical monitoring and home medical care, it is very necessary to develop a convenient and continuous blood pressure measurement method and equipment. Summary of the Invention

[0003] The present invention aims to provide a continuous blood pressure measurement method and device based on fingertip pulse waves. The invention includes fingertip pulse wave signal acquisition and feature extraction, construction of a blood pressure estimation model based on pulse wave features, embedded blood pressure calculation, and overall system programming.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] A continuous blood pressure measurement device based on fingertip pulse wave, comprising a fingertip pulse wave signal acquisition module, a blood pressure modeling module, an embedded blood pressure calculation module and a whole system design module;

[0006] The pulse wave signal acquisition module is used to acquire human fingertip pulse wave signals;

[0007] The blood pressure modeling module is used to select and optimize the pulse wave characteristics of the pulse wave time domain and frequency domain of the collected human fingertip pulse wave signal, and to optimize and train the support vector machine model for predicting blood pressure based on the pulse wave of the public database and the experimental database;

[0008] The embedded blood pressure calculation module is used to deploy the trained support vector machine model on the single-chip microcomputer, and write the input feature calculation function and model output function to complete the prediction output of blood pressure, as well as the calculation of pulse rate and blood oxygen physiological parameters;

[0009] The whole system design module is used for measuring device system program design, human-computer interaction button management, user and parameter display management, and Bluetooth remote transmission of measurement data.

[0010] A further improvement of the present invention is that the fingertip pulse wave acquisition module includes:

[0011] The pulse wave signal acquisition circuit module consists of a DCM03 photoelectric sensor and a Texas Instruments AFE4400 integrated analog front-end chip.

[0012] A further improvement of the present invention is that the DCM03 photoelectric sensor realizes fingertip contact emission of red light and infrared light and receives corresponding reflected light signals. The received fingertip reflected signals provide a data source for subsequent pulse wave extraction, blood pressure estimation and blood oxygen saturation calculation.

[0013] A further improvement of the present invention is that the AFE4400 integrated analog front-end chip realizes the driving of red light and infrared light, the reception, amplification, filtering, digital-to-analog conversion, timing sampling and data transmission of the reflected signal, and completes the signal acquisition function of the dual-path optical signal.

[0014] A further improvement of the present invention is that the blood pressure modeling module includes:

[0015] Utilizing the close relationship between blood pressure and pulse waves, the correlation between the characteristic parameters of the pulse waveform in morphology, time domain, and frequency domain and blood pressure was constructed. Then, a ranking-based feature aggregation method combining four feature optimization methods was used to compress and select the six characteristic parameters most closely related to blood pressure prediction as blood pressure estimation parameters, which were then used as input parameters for the support vector machine regression model.

[0016] The dataset required for building the support vector machine model uses the MIMIC III public dataset. According to a calibrated data selection method, data from different time periods of a total of 190 subjects were randomly selected. Among them, 100 subjects were used for feature selection, 65 for determining feature ranking, 35 for comprehensive error testing, and the remaining 90 subjects were used for model construction.

[0017] A further improvement of the present invention is that the embedded blood pressure calculation module includes:

[0018] The continuous blood pressure calculation module includes pulse wave signal preprocessing, characteristic value calculation and blood pressure value calculation, completing the functions of pulse wave signal extraction, characteristic calculation and blood pressure physiological parameter calculation;

[0019] Pulse wave signal preprocessing uses a combination of IIR and FIR low-pass filters to remove high-frequency noise and baseline drift. The high-performance IIR filter removes high-frequency noise interference, and the linear-phase FIR filter extracts low-frequency baseline drift. The baseline signal is then subtracted from the original signal to obtain a pure pulse wave signal.

[0020] The feature calculation completes the real-time calculation of pulse wave feature values ​​in two time domains and five frequency domains. Using the application program interface provided by the ARM microcontroller, the 4-byte floating-point number FFT algorithm and automatic peak detection algorithm are completed to obtain the seven characteristic parameters for blood pressure calculation.

[0021] The blood pressure value calculation completes the deployment and calculation of the embedded single-chip microcomputer of the support vector machine regression model, writes the output function of the model, and inputs the aforementioned seven characteristic parameters into the model after normalization to complete the blood pressure estimation. The blood oxygen saturation is calculated based on the dual-channel pulse wave signal, and the pulse rate is obtained by interpolation conversion of the fundamental frequency of the pulse wave spectrum.

[0022] A further improvement of the present invention is that in the whole system design module, the system program design takes signal acquisition, data processing, feature calculation, and parameter calculation as the four major tasks, and manages the eight display interfaces of user selection, parameter measurement, curve display, and historical data review of the measuring device through timed cyclic scanning of the four buttons of down, up, confirmation, and send, thereby completing the signal acquisition, measurement, display, and communication functions.

[0023] A further improvement of the present invention is that in the whole system design module, the human-computer interaction key management adopts a composite key design to realize the functions of display screen page turning, user selection, measurement start, and communication data sending.

[0024] A further improvement of the present invention is that in the whole system design module, the user and parameter display management provides multi-user management and multi-page data display management functions.

[0025] A continuous blood pressure measurement method based on fingertip pulse wave, comprising:

[0026] The pulse wave signal acquisition module acquires the pulse wave signal of the human fingertip;

[0027] The blood pressure modeling module selects and optimizes the pulse wave features in the time and frequency domains of the collected human fingertip pulse wave signals, and optimizes and trains the support vector machine model for predicting blood pressure based on pulse waves from public and experimental databases;

[0028] The embedded blood pressure calculation module deploys the trained support vector machine model on the MCU and writes the input feature calculation function and model output function to complete the prediction output of blood pressure and calculate the pulse rate and blood oxygen physiological parameters;

[0029] The whole system design module is responsible for the system program design of the measuring device, human-computer interaction button management, user and parameter display management, and Bluetooth remote transmission of measurement data.

[0030] Compared with the prior art, the present invention has at least the following beneficial technical effects:

[0031] The present invention provides a continuous blood pressure measurement method and device based on fingertip pulse waves. The method and device are implemented around PPG signal detection, machine learning analysis and blood pressure detection, and are implemented on an embedded single-chip microcomputer system. Photoplethysmography and its characteristics are used as the detection objects, and a blood pressure prediction device is used as the target. To address the problem of unclear correlation between the composition of the pulse wave feature sequence and blood pressure parameters, a variety of feature selection methods are used to study and screen data from multiple subjects at different times. A pulse wave feature combination with a strong correlation with blood pressure parameters and wide applicability is established, and a blood pressure prediction model based on a machine learning algorithm is constructed based on this. The model algorithm has been verified by public databases and self-built data, and the model algorithm has been deployed on an embedded single-chip microcomputer device. The sensor collects dual-channel photoplethysmography signals in real time, extracts its characteristic parameters and calculates with an embedded machine learning model, completing blood pressure prediction and calculation of parameters such as blood oxygen and heart rate. The present invention avoids the shortcomings of traditional electronic blood pressure monitors that use air bags and pressure sensors as detection methods in terms of wearing comfort, structure, cost control and continuous measurement. It designs and implements a set of portable continuous blood pressure detection equipment based on a single-chip microcomputer, which solves practical problems such as dual-light source fine pulse wave measurement, continuous blood pressure estimation, real-time detection and display of multiple parameters, and multi-user device sharing, providing a new blood pressure monitor for the national medical monitoring market. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0033] Figure 1 This is a block diagram of the composition of the continuous blood pressure detection device based on fingertip pulse wave of the present invention;

[0034] Figure 2 is a flow chart of the feature optimization selection method of the present invention;

[0035] Figure 3 is an error curve diagram of diastolic pressure (a) and systolic pressure (b) according to the present invention as a function of the number of features;

[0036] Figure 4 The present invention shows a Bland-Altman scatter plot (a) and a linear analysis plot (b) of diastolic pressure, and a Bland-Altman scatter plot (c) and a linear analysis plot (d) of systolic pressure;

[0037] Figure 5 This is a waveform diagram of the original pulse wave sampling signal after low-pass filtering and baseline drift removal;

[0038] Figure 6 This is a blood pressure calculation flow chart based on the support vector machine model of the embedded single-chip microcomputer system of the present invention;

[0039] Figure 7 This is the hardware block diagram of the fingertip pulse wave continuous blood pressure measurement device proposed by the present invention;

[0040] Figure 8 This is the hardware system connection diagram of the present invention

[0041] Figure 9 This is the software system programming flow chart of the present invention

[0042] Figure 10 1 is a user interface diagram of the detection device of the present invention, which includes a user selection interface (a), a physiological data acquisition interface (b), a curve drawing interface (c) and a historical data interface (d). DETAILED DESCRIPTION

[0043] Hereinafter, only certain exemplary embodiments are briefly described. As will be appreciated by those skilled in the art, the described embodiments may be modified in various ways without departing from the spirit or scope of the present invention. Therefore, the drawings and description are to be considered as illustrative in nature and not restrictive.

[0044] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like to indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0045] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0046] In the present invention, unless otherwise expressly specified or limited, terms such as "mounted," "connected," "connect," and "fixed" should be understood broadly. For example, they may refer to fixed connection, detachable connection, or integration; mechanical connection, electrical connection, or communication; direct connection or indirect connection through an intermediate medium; and internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0047] In the present invention, unless otherwise expressly specified or limited, a first feature being "above" or "below" a second feature may include the first and second features being in direct contact, or may include the first and second features being in contact not directly but through another feature between them. Furthermore, a first feature being "above," "above," and "above" a second feature may include the first feature being directly above or obliquely above the second feature, or may simply mean that the first feature is higher in level than the second feature. A first feature being "below," "below," and "below" a second feature may include the first feature being directly above or obliquely above the second feature, or may simply mean that the first feature is lower in level than the second feature.

[0048] It will be understood that when used in this specification and the appended claims, the terms “comprises” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0049] It should also be understood that the terms used in the present specification are only for the purpose of describing particular embodiments and are not intended to limit the present invention. As used in the present 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.

[0050] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0051] The accompanying drawings illustrate various schematic diagrams of structures according to embodiments disclosed herein. These figures are not drawn to scale; for clarity, some details are exaggerated and some details may be omitted. The shapes of the various regions and layers shown in the figures, as well as their relative sizes and positional relationships, are merely exemplary and may deviate in practice due to manufacturing tolerances or technical limitations. Those skilled in the art may design regions / layers with different shapes, sizes, and relative positions as needed.

[0052] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0053] Example 1

[0054] The present invention provides a continuous blood pressure measurement device based on fingertip pulse wave, which includes a fingertip pulse wave signal acquisition module, a blood pressure modeling module, an embedded blood pressure calculation module and a whole system design module;

[0055] The pulse wave signal acquisition module is used to acquire human fingertip pulse wave signals;

[0056] The blood pressure modeling module is used to select and optimize the pulse wave characteristics of the pulse wave time domain and frequency domain of the collected human fingertip pulse wave signal, and to optimize and train the support vector machine model for predicting blood pressure based on the pulse wave of the public database and the experimental database;

[0057] The embedded blood pressure calculation module is used to deploy the trained support vector machine model on the single-chip microcomputer, and write the input feature calculation function and model output function to complete the prediction output of blood pressure, as well as the calculation of pulse rate and blood oxygen physiological parameters;

[0058] The whole system design module is used for measuring device system program design, human-computer interaction button management, user and parameter display management, and Bluetooth remote transmission of measurement data.

[0059] In this embodiment, the fingertip pulse wave acquisition module includes:

[0060] The pulse wave signal acquisition circuit module consists of a DCM03 photoelectric sensor and a Texas Instruments AFE4400 integrated analog front-end chip.

[0061] The DCM03 photoelectric sensor transmits red and infrared light upon fingertip contact and receives the corresponding reflected light signals. The received fingertip reflected signals provide the data source for subsequent pulse wave extraction, blood pressure estimation, and blood oxygen saturation calculation. The AFE4400 integrated analog front-end chip implements red and infrared light driving, reflected signal reception, amplification, filtering, digital-to-analog conversion, timed sampling, and SPI interface data transmission, completing the signal acquisition function for dual-path optical signals.

[0062] In this embodiment, the blood pressure modeling module includes:

[0063] Utilizing the close relationship between blood pressure and pulse waves, the correlation between the characteristic parameters of the pulse waveform in morphology, time domain, and frequency domain and blood pressure was constructed. Then, a ranking-based feature aggregation method combining four feature optimization methods was used to compress and select the six characteristic parameters most closely related to blood pressure prediction as blood pressure estimation parameters, which were then used as input parameters for the support vector machine regression model.

[0064] The dataset required for building the support vector machine model uses the MIMIC III public dataset. According to a calibrated data selection method, data from different time periods of a total of 190 subjects were randomly selected. Among them, 100 subjects were used for feature selection, 65 for determining feature ranking, 35 for comprehensive error testing, and the remaining 90 subjects were used for model construction.

[0065] In this embodiment, the embedded blood pressure calculation module includes:

[0066] The continuous blood pressure calculation module includes pulse wave signal preprocessing, characteristic value calculation and blood pressure value calculation, completing the functions of pulse wave signal extraction, characteristic calculation and blood pressure physiological parameter calculation;

[0067] Pulse wave signal preprocessing uses a combination of IIR and FIR low-pass filters to remove high-frequency noise and baseline drift. The high-performance IIR filter removes high-frequency noise interference, and the linear-phase FIR filter extracts low-frequency baseline drift. The baseline signal is then subtracted from the original signal to obtain a pure pulse wave signal.

[0068] The feature calculation completes the real-time calculation of pulse wave feature values ​​in two time domains and five frequency domains. Using the application program interface provided by the ARM microcontroller, the 4-byte floating-point number FFT algorithm and automatic peak detection algorithm are completed to obtain the seven characteristic parameters for blood pressure calculation.

[0069] The blood pressure value calculation completes the deployment and calculation of the embedded single-chip microcomputer of the support vector machine regression model, writes the output function of the model, and inputs the aforementioned seven characteristic parameters into the model after normalization to complete the blood pressure estimation. The blood oxygen saturation is calculated based on the dual-channel pulse wave signal, and the pulse rate is obtained by interpolation conversion of the fundamental frequency of the pulse wave spectrum.

[0070] In this embodiment, in the whole system design module, the system program design is to take signal acquisition, data processing, feature calculation, and parameter calculation as the four major tasks. Through the timed loop scanning of the four buttons of down, up, confirm, and send, the system manages the eight display interfaces of the measuring device, including user selection, parameter measurement, curve display, and historical data viewing, and completes the functions of signal acquisition, measurement, display, and communication.

[0071] The human-computer interaction key management adopts a duplex key design to realize functions such as display screen page turning, user selection, measurement start, communication data sending, etc.

[0072] User and parameter display management provides multi-user management and multi-page data display management functions.

[0073] Example 2

[0074] The present invention provides a continuous blood pressure measurement method based on fingertip pulse waves, which includes fingertip pulse wave signal acquisition and feature extraction, construction of a blood pressure estimation model based on pulse wave features, embedded blood pressure calculation and whole system programming.

[0075] Figure 1 This is a block diagram of the system components of the blood pressure measurement device proposed in the present invention. First, 78 characteristic parameters of the pulse wave are extracted in the morphological, time, frequency, wavelet, and Hilbert transform domains. Four feature selection methods are then used to optimize and obtain six characteristic parameters consistent with blood pressure estimation. These six characteristic parameters are then used to construct a blood pressure estimation model using different machine learning methods. The optimal support vector machine-based regression model is then obtained as the device's blood pressure estimation model. Finally, the microcontroller-based pulse wave acquisition, preprocessing, feature extraction, pulse rate blood sample calculation, and machine learning-based blood pressure calculation based on the six features are completed. The 30-second curves and averages of heart rate, blood oxygen, diastolic pressure, and systolic pressure are then displayed on an LCD screen.

[0076] Figure 2 This is a flow chart of the optimization selection method for pulse wave feature extraction. 78-dimensional features are extracted from pulse wave data of multiple subjects. Four feature parameter optimization methods are used to obtain four parameter contribution rankings. Then, feature aggregation based on ranking is obtained. Finally, the following is obtained: Figure 3 The error curve of the blood pressure calculation result with different characteristic parameter numbers is obtained. Taking into account the embedded system transplantation and fast calculation, the present invention selects Figure 3 The first six blood pressure calculation characteristic parameters in the model are used as the input parameters of the model building: second harmonic frequency, second harmonic energy percentage, peak time interval, the ratio of the energy spectrum of 0.25-10Hz to 0.25-20Hz, the ratio of the energy spectrum of 0.25-3.5Hz to 0.25-20Hz, and the ratio of the energy spectrum of 0.25-3.5Hz to 0.25-10Hz.

[0077] The proposed support vector machine blood pressure calculation method based on pulse wave features was validated using 2,250 data sets from 90 subjects in the publicly available MIMIC database. The results showed that the mean absolute error (MAE) and standard deviation (STD) for diastolic blood pressure (DBP) were 3.65 mmHg and 6.09 mmHg, respectively, and the MAE and STD for systolic blood pressure (SBP) were 4.13 mmHg and 6.34 mmHg, respectively. Both met the Association for the Advancement of Medical Instrumentation (AAMI) standards of less than 5 mmHg for MAE and 8 mmHg for STD, respectively, and also met the British Hypertension Society (BHS) Grade A criteria. Figure 4This is the Bland-Altman scatter plot and correlation analysis of blood pressure estimation. The correlation coefficients of DBP and SBP in the figure are 0.84 and 0.96, respectively, indicating that the model estimation has a strong correlation. At the same time, most of the prediction results of DBP and SBP are within the 95% consistency interval of the error mean ±1.96SD, and the error mean of SBP and DBP are close to 0, indicating that the model estimation has good consistency.

[0078] The present invention simultaneously collects the red spectrum and infrared spectrum reflectance PPG signals of the human middle finger. The sensor sampling rate is 125Hz, and the data processing window is 2048 sampling points. The PPG signal preprocessing mainly includes high-frequency noise filtering and baseline drift removal. High-frequency noise filtering is completed by using a high-order and phase-insensitive infinite impulse response IIR filter. Baseline drift filtering is completed by using a linear phase finite impulse response FIR filter to extract the low-frequency baseline signal, and then subtract the baseline signal from the original signal. The PPG preprocessing result is as follows: Figure 5 shown.

[0079] Blood oxygen saturation (SpO2) refers to the degree of binding of oxygen to hemoglobin in the blood and is an important indicator of the body's blood oxygen content. The present invention calculates blood oxygen saturation using the following calibration fitting formula:

[0080] SpO2= -18.653R 2 + 5.701R + 102.1449 (1)

[0081]

[0082] Where: R is the ratio of red light to infrared light DC to AC. Corresponding to the AC component of red light, Corresponding to the DC component of red light, Corresponding to the AC component of infrared, The pulse wave acquisition module in the system contains two light sources, red and infrared, which flash alternately to obtain a dual-channel PPG. The calculation of the DC and AC components extracts the peaks, and the troughs between the two sets of peaks are extracted. The trough value is the DC component, and the difference between the two is the AC component.

[0083] The present invention adopts the support vector machine for regression (SVR) machine learning model to complete the blood pressure estimation method based on pulse wave characteristics to complete blood pressure detection, and completes the deployment calculation of the model decision function on the single chip microcomputer STM32H573. The overall blood pressure prediction algorithm process design is as follows Figure 6To verify the accuracy of the decision function, 100 sets of identical pulse wave features were selected and input into the STM32 decision function and the prediction function in MATLAB. The ratios of the two methods' prediction results for diastolic and systolic blood pressure were 100.015% and 100.009%, respectively, indicating that the prediction results were similar and that the deployment on the microcontroller had good reliability.

[0084] Figure 7 The hardware system designed by the present invention mainly includes a main control CPU with STM32H753 as the core, which performs control functions such as data acquisition, calculation, display, and transmission; 2MB of FLASH and 1MB of RAM memory for storing system programs and data; a 4.3-inch capacitive touch screen TFT-LCD with a resolution of 800*480 for providing parameter display and human-computer interaction; a dual-input pulse wave acquisition module with the biosensor AFE4400 as the core, which performs real-time acquisition of red light and infrared light emission and reflection reception data; a Bluetooth module mainly for long-distance wireless transmission of blood pressure, heart rate, and blood oxygen parameters; a single 5V power input, and internal voltage division and voltage regulation of different levels to meet the different voltage requirements of the system. Figure 8 It is a physical connection diagram of each part.

[0085] Figure 9 This is a flowchart of the software system programming of the present invention. After the system is powered on and initialized, it mainly realizes the system functions such as user registration, parameter measurement, test browsing, and historical data recording through the three identification variables KEY, SCREEN_num, and USER_name under the control of the four buttons down, up, congfirm, and trans.

[0086] Example 3

[0087] This method employs a blood pressure estimation method based on PPG characteristic parameters. It first performs multidimensional feature extraction, employing various feature selection methods to generate multiple feature rankings. Feature aggregation methods are then used to determine pulse wave feature subsequences required for subsequent research. Based on the determined pulse wave characteristics, feature sequences are extracted from the PPG signal, and machine learning is used to construct a blood pressure prediction model. Finally, based on the application scenario of an embedded system, the model's time and space complexity and portability are ensured to determine the final blood pressure prediction model for subsequent embedded system implementation.

[0088] To verify the system's measurement accuracy, data were collected from six healthy subjects aged between 20 and 30, including one female with hypotension. Data were collected over three days, with five sets collected daily, for a total of 90 sets. Blood pressure was measured on the subjects' right arms using a YXY-61 electronic blood pressure monitor, with systolic and diastolic pressures recorded as label values. An STM32 was used to collect pulse waves from the subjects' right index fingers and predict blood pressure, recording systolic and diastolic pressures as predicted values. The experimental results are shown in Table 1.

[0089] Table 1 Measurement results error

[0090]

[0091] The detection results of the finger oximeter of 5 subjects were used as labels and tested with the pulse wave using the fitting formula (1). The results are shown in Table 2.

[0092] Table 1 SPO2 experimental results

[0093]

[0094] according to Figure 7 Hardware system block diagram and Figure 8 The hardware connection mode constitutes the hardware detection platform of the present invention. Figure 9 Design corresponding system software program to complete user's convenient control of the equipment through 4 touch buttons, realize functions such as user registration, parameter measurement, test browsing, historical data recording, etc. Figure 10 The system provides the power-on interface, measurement interface, parameter waveform display interface and historical data viewing interface.

[0095] On the user startup page, press the up / down buttons to cycle through the five users, and then press the confirm button to go to the next measurement page.

[0096] On the measurement page, the user confirms on the home page and turns to the measurement page. Press the confirmation button to turn on the measurement function. The screen countdown completes 30 seconds of fingertip pulse wave data collection. After less than 1 second of calculation, the human physiological parameters are displayed. The measurement parameters include systolic blood pressure, diastolic blood pressure, mean blood pressure, blood oxygen, pulse rate and other parameters.

[0097] On the curve display page, select the page using the up / down buttons to display the change curves of blood pressure, blood oxygen, and pulse rate within 30 seconds.

[0098] On the historical data page, the historical data of 5 users is displayed, with a total of 20 sets of data for each user recorded and saved in a loop.

[0099] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, from all points of view, the embodiments should be regarded as illustrative and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes that fall within the meaning and range of equivalents of the claims are included in the present invention. Any reference signs in the claims should not be construed as limiting the claim to which they relate.

[0100] In addition, it should be understood that although this specification describes the embodiments, not every embodiment contains only one independent technical solution. This description is for clarity only. Those skilled in the art should consider the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for the purpose of illustrating the technical concept of the present invention and cannot be used to limit the scope of protection of the present invention. Any changes made based on the technical solution in accordance with the technical concept proposed by the present invention fall within the scope of protection of the claims of the present invention.

Claims

1. A continuous blood pressure measurement device based on fingertip pulse wave, characterized in that: It includes fingertip pulse wave signal acquisition module, blood pressure modeling module, embedded blood pressure calculation module and whole system design module; The pulse wave signal acquisition module is used to acquire human fingertip pulse wave signals; The blood pressure modeling module is used to select and optimize the pulse wave characteristics of the pulse wave time domain and frequency domain of the collected human fingertip pulse wave signal, and to optimize and train the support vector machine model for predicting blood pressure based on the pulse wave of the public database and the experimental database; The embedded blood pressure calculation module is used to deploy the trained support vector machine model on the single-chip microcomputer, and write the input feature calculation function and model output function to complete the prediction output of blood pressure, as well as the calculation of pulse rate and blood oxygen physiological parameters; The whole system design module is used for measuring device system program design, human-computer interaction button management, user and parameter display management, and Bluetooth remote transmission of measurement data.

2. The continuous blood pressure measurement device based on fingertip pulse wave according to claim 1, characterized in that: Fingertip pulse wave acquisition module, including: The pulse wave signal acquisition circuit module consists of a DCM03 photoelectric sensor and a Texas Instruments AFE4400 integrated analog front-end chip.

3. The continuous blood pressure measurement device based on fingertip pulse wave according to claim 2, characterized in that: The DCM03 photoelectric sensor emits red and infrared light upon fingertip contact and receives corresponding reflected light signals. The received fingertip reflected signals provide a data source for subsequent pulse wave extraction, blood pressure estimation, and blood oxygen saturation calculation.

4. The continuous blood pressure measurement device based on fingertip pulse wave according to claim 2, characterized in that: The AFE4400 integrated analog front-end chip realizes the driving of red and infrared light, the reception, amplification, filtering, digital-to-analog conversion, timing sampling and data transmission of the reflected signal, and completes the signal acquisition function of the dual-channel optical signal.

5. The continuous blood pressure measurement device based on fingertip pulse wave according to claim 1, characterized in that: Blood pressure modeling module, including: Utilizing the close relationship between blood pressure and pulse waves, the correlation between the characteristic parameters of the pulse waveform in morphology, time domain, and frequency domain and blood pressure was constructed. Then, a ranking-based feature aggregation method combining four feature optimization methods was used to compress and select the six characteristic parameters most closely related to blood pressure prediction as blood pressure estimation parameters, which were then used as input parameters for the support vector machine regression model. The dataset required for building the support vector machine model uses the MIMIC III public dataset. According to a calibrated data selection method, data from different time periods of a total of 190 subjects were randomly selected. Among them, 100 subjects were used for feature selection, 65 for determining feature ranking, 35 for comprehensive error testing, and the remaining 90 subjects were used for model construction.

6. The continuous blood pressure measurement device based on fingertip pulse wave according to claim 1, characterized in that: Embedded blood pressure calculation module, including: The continuous blood pressure calculation module includes pulse wave signal preprocessing, characteristic value calculation and blood pressure value calculation, completing the functions of pulse wave signal extraction, characteristic calculation and blood pressure physiological parameter calculation; Pulse wave signal preprocessing uses a combination of IIR and FIR low-pass filters to remove high-frequency noise and baseline drift. The high-performance IIR filter removes high-frequency noise interference, and the linear-phase FIR filter extracts low-frequency baseline drift. The baseline signal is then subtracted from the original signal to obtain a pure pulse wave signal. The feature calculation completes the real-time calculation of pulse wave feature values ​​in two time domains and five frequency domains. Using the application program interface provided by the ARM microcontroller, the 4-byte floating-point number FFT algorithm and automatic peak detection algorithm are completed to obtain the seven characteristic parameters for blood pressure calculation. The blood pressure value calculation completes the deployment and calculation of the embedded single-chip microcomputer of the support vector machine regression model, writes the output function of the model, and inputs the aforementioned seven characteristic parameters into the model after normalization to complete the blood pressure estimation. The blood oxygen saturation is calculated based on the dual-channel pulse wave signal, and the pulse rate is obtained by interpolation conversion of the fundamental frequency of the pulse wave spectrum.

7. The continuous blood pressure measurement device based on fingertip pulse wave according to claim 1, characterized in that: In the overall system design module, the system program design takes signal acquisition, data processing, feature calculation, and parameter calculation as the four major tasks. Through timed cyclic scanning of the four buttons of down, up, confirm, and send, it manages the eight display interfaces of user selection, parameter measurement, curve display, and historical data review of the measuring device to complete the signal acquisition, measurement, display, and communication functions.

8. The continuous blood pressure measurement device based on fingertip pulse wave according to claim 1, characterized in that: In the whole system design module, the human-computer interaction button management adopts a duplex button design to realize the functions of display screen page turning, user selection, measurement start, and communication data sending.

9. The continuous blood pressure measurement device based on fingertip pulse wave according to claim 1, characterized in that: In the whole system design module, user and parameter display management provides multi-user management and multi-page data display management functions.

10. A continuous blood pressure measurement method based on fingertip pulse wave, characterized in that: include: The pulse wave signal acquisition module acquires the pulse wave signal of the human fingertip; The blood pressure modeling module selects and optimizes the pulse wave features in the time and frequency domains of the collected human fingertip pulse wave signals, and optimizes and trains the support vector machine model for predicting blood pressure based on pulse waves from public and experimental databases; The embedded blood pressure calculation module deploys the trained support vector machine model on the MCU and writes the input feature calculation function and model output function to complete the prediction output of blood pressure and calculate the pulse rate and blood oxygen physiological parameters; The whole system design module is responsible for the system program design of the measuring device, human-computer interaction button management, user and parameter display management, and Bluetooth remote transmission of measurement data.

Images