Continuous arterial blood pressure waveform detection system and equipment based on single-channel fingertip PPG

Abstract

The present invention discloses a continuous arterial blood pressure waveform detection system and equipment based on single-channel fingertip PPG, and relates to the field of blood pressure detection devices. The detection system includes several functional modules including a calibration module, a feature input module, a signal processing module and a signal processing module. The system can collect the PPG signal of the subject to be tested and pre-process the collected PPG signal, and then call the pre-established elastic cavity model and Tube-Load model with different population characteristics to form a hybrid model, and input the PPG signal of the subject to the established hybrid model to obtain the real-time continuous blood pressure waveform of the subject. The method and device of the present invention can better overcome the shortcomings of the existing blood pressure detection model, and can realize real-time detection of blood pressure waveform.

Description

Technical Field

[0001] The present invention relates to the field of physiological modeling, and in particular to a hybrid continuous blood pressure waveform detection system and device based on a single-channel fingertip photoplethysmography (PPG) waveform. Background Art

[0002] Blood pressure refers to the lateral pressure exerted on the vessel wall per unit area when blood flows in the blood vessels. It is the driving force that pushes blood to flow in the blood vessels. Measuring arterial blood pressure (ABP) can significantly improve cardiovascular health monitoring. Accurate and timely blood pressure measurement is essential for preventing hypertension and related cardiovascular diseases.

[0003] At present, the prevalence of hypertension is high, the damage is great, and the patient population is relatively young. The existing gold standard for blood pressure detection is relatively harmful to the human body; and some common non-invasive blood pressure detection devices are still based on cuffs, which limits the monitoring of continuous blood pressure; most of the devices that can perform continuous blood pressure detection now use two or more sensors, which are not only uncomfortable but also not suitable for wear. In the continuous blood pressure detection model construction method of the present invention, most of the other inventions that use the elastic cavity model are currently for obtaining the relevant hemodynamic characteristics of the blood vessels; most of the other inventions that use the Tube-Load model are for obtaining the proximal pressure waveform from the distal pressure waveform. There has been no research on the construction method of the blood pressure detection model similar to the present invention, that is, combining the elastic cavity model and the Tube-Load model to construct the peripheral vascular transfer function model of the corresponding part.

[0004] Therefore, the present invention provides a method for obtaining radial artery dynamic blood pressure waveform using only fingertip PPG signal. First, fingertip PPG collection is more convenient. Compared with fingertip PPG measurement, pulse wave measurement at the radial artery is more complicated, and the radial artery needs to be accurately located. Moreover, compared with wearing sensors on fingers, wearing sensors on wrists will cause more discomfort to the human body. Moreover, compared with the commonly used clinical puncture to obtain dynamic blood pressure values, this invention reduces unnecessary pain for patients and has great research value and application prospects. Summary of the invention

[0005] The present invention provides a continuous arterial blood pressure waveform detection system and equipment based on single-channel fingertip PPG, which realizes the continuous monitoring of the finger artery PPG signal to the radial artery blood pressure waveform. The present invention adopts the elastic cavity model and the Tube-Load model for mathematical modeling, so that the parameters in the formed transfer function model have a certain physiological explanation; the present invention divides the subjects into groups based on different population characteristics, and obtains the optimal value range of each characteristic population model parameter, which is convenient for studying the influence of parameter changes on the blood pressure waveform, making dynamic blood pressure detection more convenient and accurate, which not only meets the clinical needs of patients with cardiovascular diseases, but also plays a certain monitoring role for the normal population, and has very important practical significance for human life and health research.

[0006] The technical solutions specifically adopted in the present invention are as follows:

[0007] In a first aspect, the present invention provides a continuous arterial blood pressure waveform detection system based on a single-channel fingertip PPG, comprising:

[0008] The calibration module is used to obtain the fingertip PPG signal, reference blood pressure signal and cardiac output CO of the subject collected by the external device during the calibration phase, and calibrate the blood viscosity resistance R, vascular compliance C and fingertip characteristic impedance Z of the subject based on the fingertip PPG signal and reference blood pressure signal. c 1;

[0009] The feature input module is used for the subject to be tested to input his / her personal features, and obtain the feature coefficients of the Tube-Load model adapted to the subject according to the mapping relationship between the pre-stored personal features and the feature coefficients of the Tube-Load model; the personal features include age, gender and the presence or absence of abnormal cardiovascular status;

[0010] A signal processing module, used for acquiring the fingertip PPG signal of the subject to be tested collected in real time by an external device, and after performing signal preprocessing on the fingertip PPG signal, using the cardiac output CO to calibrate the fingertip PPG signal into the total volume pulse blood flow Q(t) of the subject to be tested;

[0011] The blood pressure waveform output module is used to input the total volume pulse blood flow Q(t) of the subject to be tested obtained in the signal processing module into a hybrid model composed of an elastic cavity model and a Tube-Load model, and firstly convert the total volume pulse blood flow Q(t) into a pressure waveform P(t) of the fingertip pulse wave by the elastic cavity model, and then convert the pressure waveform P(t) of the fingertip pulse wave into an arterial blood pressure waveform ABP(t) of the radial artery by the Tube-Load model; wherein the blood viscosity resistance R, vascular compliance C and fingertip characteristic impedance Z used in the transfer function of the elastic cavity model are c1 is obtained by the calibration module, and the characteristic coefficients used in the transfer function of the Tube-Load model are obtained by the characteristic input module.

[0012] As a preferred embodiment of the first aspect, the hybrid model is composed of an elastic cavity model and a Tube-Load model, wherein the model input is the total volume pulse blood flow Q(t) of the subject to be tested, and the final output is the arterial blood pressure waveform ABP(t) of the radial artery of the subject to be tested, expressed as:

[0013] ABP(t)=Q(t)*g(t)*h(t)

[0014] Wherein: the symbol * represents the convolution operation, g(t) is the result of inverse Laplace transform of the s-domain transfer function G(s) of the elastic cavity model, and h(t) is the result of inverse Laplace transform of the w-domain transfer function H(w) of the Tube-Load model;

[0015] The transfer function G(s) of the elastic cavity model is composed of blood viscosity resistance R, vascular compliance C and fingertip characteristic impedance Z c 1 is determined, and the expression is as follows:

[0016]

[0017] The transfer function H(w) of the Tube-Load model is determined by the reflection coefficient T(w) and the time Δt for the pulse wave to propagate from the radial artery to the fingertip, and is expressed as follows:

[0018]

[0019] The reflection coefficient A and B are the characteristic coefficients.

[0020] As a preferred embodiment of the first aspect, the calibration program in the calibration module is run regularly. Each time the calibration program is run, the fingertip PPG signal, the reference blood pressure signal and the cardiac output CO of the subject to be tested are collected through an external device, and then the systolic pressure P is obtained from the obtained reference blood pressure signal. s , diastolic blood pressure d and mean arterial pressure P a , the average volume pulse blood flow Q is obtained from the fingertip PPG signal obtained during the calibration phase m and maximum volume pulse blood flow Q max , obtain the stroke volume SV through the cardiac output CO, and then use the formula Obtain the blood viscosity resistance R of the subject to be tested, through the formula Obtain the vascular compliance C of the subject to be tested, through the formula Get the characteristic impedance Z of the fingertip of the person being tested c1 value, thereby updating the cardiac output CO value, blood viscosity resistance R value, vascular compliance C value and fingertip characteristic impedance Z value of the subject to be tested stored in the system c 1 value.

[0021] As a preferred embodiment of the first aspect, in the feature input module, the mapping relationship between the stored personal features and the Tube-Load model feature coefficients is obtained by an adaptive system identification method, and the specific identification method is:

[0022] First, the personal characteristics are divided into multiple categories according to different combinations of three indicators: age, gender, and presence or absence of abnormal cardiovascular status;

[0023] Then, the forward pressure wave Pf(t) and the reflected pressure wave Pb(t) are used as the input and output of the system identification, respectively, and the reflection coefficient is calculated using the sample data of each category. Regression is performed to determine the values ​​of characteristic coefficients A and B corresponding to each category, thereby establishing a mapping relationship between personal characteristic categories and characteristic coefficients of the Tube-Load model.

[0024] As a preferred embodiment of the first aspect above, in the signal processing module, preprocessing of the fingertip PPG signal includes: eliminating incomplete signals and abnormal signals, using wavelet transform to process baseline drift caused by breathing, external light source and movement factors in the signal, and using a bandpass filter to process high-frequency interference and glitches caused by other electromagnetic factors in the signal.

[0025] In a second aspect, the present invention provides a computer electronic device comprising a memory and a processor;

[0026] The memory is used to store computer programs;

[0027] The processor is used to implement the hybrid continuous blood pressure detection system described in any one of the first aspects above when executing the computer program.

[0028] In a third aspect, the present invention provides a hybrid continuous arterial blood pressure waveform detection device based on a single-channel fingertip PPG, characterized in that it includes a paired PPG signal acquisition device, a blood pressure signal acquisition device, a cardiac output acquisition device, and an arterial blood pressure detection device;

[0029] The PPG signal acquisition device is used to acquire the fingertip PPG signal of the subject to be tested after receiving the calibration instruction or blood pressure detection instruction issued by the arterial blood pressure detection device;

[0030] The blood pressure signal acquisition device is used to acquire a reference blood pressure signal of the subject to be measured after receiving a calibration instruction issued by the arterial blood pressure detection device;

[0031] The cardiac output acquisition device is used to acquire the cardiac output CO of the subject to be tested after receiving the calibration instruction issued by the arterial blood pressure detection device;

[0032] The arterial blood pressure detection device is used to send calibration instructions and blood pressure detection instructions to the remaining acquisition devices, and at the same time receive signal data uploaded by each acquisition device; and based on the received data, call the hybrid continuous blood pressure detection system as described in any one of the first aspects above to output the arterial blood pressure waveform ABP(t) of the radial artery of the person to be tested.

[0033] As a preferred embodiment of the third aspect, the arterial blood pressure detection device is a mobile device or a host computer, and the hybrid continuous blood pressure detection system exists in the form of software, and provides the user with personal feature input and arterial blood pressure waveform display functions through a UI interface.

[0034] As a preferred embodiment of the third aspect, the PPG signal acquisition device is a wearable device worn on the end of a finger.

[0035] As a preferred embodiment of the third aspect, the PPG signal acquisition device, the blood pressure signal acquisition device, and the cardiac output acquisition device are respectively paired and connected with the arterial blood pressure detection device through a Bluetooth module to perform two-way transmission of data and instructions.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] On the basis of continuous blood pressure detection, the present invention takes into account the role of potential damping factors and the propagation and reflection of waves in blood circulation, and combines the elastic cavity model with the Tube-Load model into a new hybrid model, which not only reflects the physiological characteristics of the cardiovascular system but also realizes the detection of continuous blood pressure waveforms. The present invention obtains the optimal model parameters of different population characteristics in advance through a system identification method, thereby realizing personalized detection of the model. In addition, the hybrid model of the present invention also shows good blood pressure trend tracking performance under the condition of blood pressure fluctuations after initial parameter calibration based on its personalized characteristics. The most common continuous blood pressure detection method based on PTT not only increases the complexity of the task but also has the problem of time synchronization. The present invention directly estimates the continuous blood pressure waveform of the radial artery through the PPG signal detected by the fingertips, and realizes the synchronous and imperceptible monitoring of flow and blood pressure within the clinically acceptable accuracy range. It is an improved continuous blood pressure waveform detection scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic diagram of a single-cycle volume pulse wave signal in one embodiment of the present invention.

[0039] Figure 2 It is a schematic diagram of a single-cycle ABP signal in one embodiment of the present invention.

[0040] Figure 3 It is a schematic diagram of the module composition of a continuous arterial blood pressure waveform detection system based on single-channel fingertip PPG in one embodiment of the present invention.

[0041] Figure 4 It is a schematic diagram of the principle of a continuous arterial blood pressure waveform detection system based on single-channel fingertip PPG in one embodiment of the present invention.

[0042] Figure 5 is the estimated ABP signal and the measured ABP signal in one embodiment of the present invention

[0043] Figure 6 It is a schematic diagram of a computer electronic device in one embodiment of the present invention.

[0044] Figure 7 2 is a diagram of a PPG waveform acquisition device in one embodiment of the present invention.

[0045] Figure 8 1 is a flowchart of a PPG waveform collection process in one embodiment of the present invention.

[0046] Fig. 9 This is a control flow chart of the APP side in one embodiment of the present invention.

[0047] Fig.10 It is a Bluetooth Socket communication flow chart in one embodiment of the present invention.

[0048] Fig.11 This is an APP interface diagram in one embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] To facilitate understanding, the relevant concepts involved in this application are first explained.

[0051] (1) Pulse wave

[0052] The pulse wave is formed by the heart's pulsation or vibration propagating along the arteries and blood flow to the periphery during the blood circulation process. Its density and viscosity are closely related to the elasticity, caliber and thickness of the arterial wall. The characteristic changes of the pulse wave waveform are an important basis for evaluating the physiological and pathological state of the human cardiovascular system. Therefore, people always try to extract various physiological and pathological information from the changes in the pulse wave waveform.

[0053] When the pulse wave propagates in the arteries, the blood pressure, velocity and blood vessel diameter at the place where it arrives pulsate at the same time; pulse waves are usually divided into pressure pulse waves and volume pulse waves, and their propagation velocities are the same but the waveforms are different; pressure pulse waves refer to the propagation process of blood pressure pulsation. Since the pulsation curve of blood vessel diameter over time is approximately equal to the waveform of pressure pulsation inside the blood vessel, various pressure sensors such as strain gauge and piezoelectric are often used to detect the curve of blood vessel diameter over time to obtain non-invasive recording of blood pressure waveform in arterial vessels; volume pulse waves reflect the propagation process of blood flow pulsation. The pulsation change of blood flow reflects many important cardiovascular information such as cardiac function, blood flow, peripheral blood vessels and microcirculation, which can generally be non-invasively detected at the fingertips of the human body through photoelectric capacitance pulse sensors;

[0054] like Figure 1 As shown in Figure 1, it is a volume pulse wave signal of a cycle, which is divided into systolic waveform and diastolic waveform. The highest point Q max and the lowest point Q min are the maximum and minimum values ​​of the volume pulse blood flow curve, Q m Represents the average volume pulse blood flow; the characteristics of the volume pulse wave signal are that the main wave amplitude is low, the tidal wave is not obvious, the dicrotic wave is low, and the rising and falling edges of the waveform are slow. The present invention mainly uses the volume pulse wave, and establishes the relationship between the volume pulse wave and the arterial blood pressure waveform by establishing a mathematical model based on the characteristics of different populations; then, through the volume pulse wave acquisition device, the established hybrid continuous blood pressure detection model and the specific transmission protocol are used to display the arterial blood pressure waveform of the person to be tested on the mobile phone APP.

[0055] (2) Blood pressure

[0056] Blood pressure refers to the lateral pressure exerted on the wall of a vessel per unit area when blood flows in the blood vessels. It is the driving force for blood to flow in the blood vessels. In different blood vessels, it is called arterial blood pressure, capillary pressure and venous blood pressure respectively. The blood pressure usually refers to the arterial blood pressure of the systemic circulation. The factors affecting arterial blood pressure mainly include stroke volume, peripheral resistance, heart rate, elasticity of the aorta and large artery walls, circulating blood volume and vascular capacity. Invasive measurement and non-invasive measurement are the two major methods of blood pressure measurement at present. The invasive measurement method is to insert a catheter connected to a pressure sensor directly into the large artery or heart through the skin to detect blood pressure signals, and can obtain dynamic blood pressure waveforms. Since this method is directly measured inside the blood vessel, it avoids interference from other factors, and the measurement data is the most accurate, which can be used as the gold standard for measuring blood pressure; non-invasive measurement of blood pressure is to indirectly obtain blood pressure by detecting the pulsation of the superficial artery wall on the body surface and the change of blood volume in the blood vessel. Non-invasive measurement methods are divided into two categories: intermittent measurement and continuous measurement. Auscultation and oscillometric methods are intermittent measurement methods. Since a cuff device is required, it may cause discomfort to the person being measured. Pulse wave velocity method, pulse wave parameter measurement method and pulse transmission time method are continuous measurement methods. However, most of these methods use more than two sensors, the devices are complex, and there is a problem of time synchronization.

[0057] like Figure 2 As shown, it is a periodic arterial blood pressure signal, the highest point P s and the lowest point P d are systolic and diastolic blood pressure, P a Represents mean pressure; usually, arterial blood pressure signals are obtained through puncture, and the selected sites are usually radial artery, brachial artery, etc., which are somewhat traumatic to the human body. Therefore, under the condition of a certain body position, the model constructed by the present invention is used to convert the finger artery volume pulse wave signal into a radial artery arterial blood pressure signal. Compared with the invasive puncture method for obtaining arterial blood pressure signals, this research has far-reaching research significance. The present invention also writes the arterial blood pressure signal as an ABP signal. The above-mentioned constant body position refers to lying flat measurement or keeping the measurement site and the heart at the same horizontal line.

[0058] (3) Elastic cavity model

[0059] The elastic cavity model is a simplified model for analyzing the blood flow characteristics of the cardiovascular system. This model regards a part of the cardiovascular system as an elastic cavity and assumes that the blood flow parameters are equal everywhere in it, so as to simulate and analyze the cardiovascular system. In terms of blood pressure detection, the elastic cavity model mainly studies the key parameters of the arterial system such as R, C, SV by establishing an equivalent circuit to simulate the pulse waveform; analyzes the impact of parameter changes on blood pressure calculation; and describes the relationship between systolic blood pressure and diastolic blood pressure inside the blood vessel. When the elastic cavity model simulates the relationship between various parameters in the blood flow in the blood vessels by establishing an equivalent circuit, it is necessary to establish an analogy relationship and unit conversion relationship between the hemodynamic parameters and the electrical parameters. The blood pressure difference drives the blood flow, and the voltage difference drives the current flow, so the blood pressure is analogous to the voltage; the blood flow flowing into the bifurcation point is equal to the sum of the blood flow flowing out of the bifurcation point, and the current also satisfies a similar law, so the blood flow is analogous to the current; the resistance simulates the viscous resistance that causes the blood pressure to drop, so the blood viscosity resistance is analogous to the resistance; the inductance simulates the blood inertia that hinders the change of blood flow, so the blood inertia is analogous to the inductance; the capacitor is used to simulate the expansion and contraction of the blood vessels during the pulsation process, so the blood vessel compliance is analogous to the capacitor. In the present invention, the radial artery of the upper arm to the finger artery is an important part of the microcirculation in the blood circulation system. A special analogy between the hemodynamic parameters and the electrical parameters is formed between the radial artery and the finger artery, which links the pressure and flow of the blood vessel outflow domain at the fingertips, so it is necessary to use the Poiseuille law of circular tube laminar flow The flow rate Q is proportional to the pressure drop ΔP and is also proportional to the fourth power of the radius a of the tube. An equivalent circuit of the elastic cavity model is established in the microcirculation.

[0060] (4) System Identification Method

[0061] The system identification method uses the system identification technology in control theory to model the blood circulation system, that is, the entire cardiovascular system is regarded as a nonlinear black box, and the internal details are not investigated, but the functional characteristics of the cardiovascular system are studied by external observation and input and output information. The system identification method mainly establishes the transfer function between the input and output points of the arterial system. The present invention uses an adaptive system identification method to solve the values ​​of A and B in the reflection coefficient of the Tube-Load model, avoiding errors caused by physiological changes and solving the problem of model personalization.

[0062] Based on the above definition of basic concepts, the present invention will specifically describe a method for constructing a hybrid continuous arterial blood pressure waveform detection model based on single-channel fingertip PPG.

[0063] In a preferred embodiment of the present invention, a continuous arterial blood pressure waveform detection system based on a single-channel fingertip PPG is provided, which takes into account the influence of potential damping factors in blood microcirculation and the propagation and reflection phenomena of waves, and realizes continuous monitoring of radial artery blood pressure waveform with higher accuracy. Figure 3 As shown, the detection system includes several functional modules including a calibration module, a feature input module, a signal processing module and a signal processing module. The specific functions and usage of each module are described in detail below.

[0064] The calibration module is used to obtain the fingertip PPG signal, reference blood pressure signal and cardiac output CO of the subject collected by the external device during the calibration phase, and calibrate the blood viscosity resistance R, vascular compliance C and fingertip characteristic impedance Z of the subject based on the fingertip PPG signal and reference blood pressure signal. c 1.

[0065] The feature input module is used for the subject to be tested to input his / her personal features, and obtain the feature coefficients of the Tube-Load model adapted to the subject according to the mapping relationship between the pre-stored personal features and the feature coefficients of the Tube-Load model; the personal features include three indicators: age, gender and the presence or absence of abnormal cardiovascular status.

[0066] The signal processing module is used to obtain the fingertip PPG signal of the subject to be tested collected in real time by an external device, and after signal preprocessing, the fingertip PPG signal is calibrated as the total volume pulse blood flow Q(t) of the subject to be tested using the cardiac output.

[0067] The blood pressure waveform output module is used to input the total volume pulse blood flow Q(t) of the subject obtained in the signal processing module into a hybrid model composed of an elastic cavity model and a Tube-Load model. The elastic cavity model first converts the total volume pulse blood flow Q(t) into a pressure waveform P(t) of the fingertip pulse wave, and then the Tube-Load model converts the pressure waveform P(t) of the fingertip pulse wave into an arterial blood pressure waveform ABP(t) of the radial artery.

[0068] The hybrid model in the above-mentioned blood pressure waveform output module is described in detail below.

[0069] like Figure 4As shown, a schematic diagram of the principle of obtaining the ABP signal through the PPG signal in the hybrid model is shown, which includes two processes, namely, after the overall volume pulse blood flow Q(t) is obtained by calibrating the fingertip PPG signal, the overall volume pulse blood flow Q(t) is first converted into the pressure waveform P(t) of the fingertip pulse wave by the elastic cavity model, and then the pressure waveform P(t) of the fingertip pulse wave is converted into the arterial blood pressure waveform ABP(t) of the radial artery by the Tube-Load model. Among them, the hybrid model in the above-mentioned blood pressure waveform output module is closely related to the personal characteristics of the subject and the parameters obtained in the calibration stage. Specifically, the blood viscosity resistance R, vascular compliance C and fingertip characteristic impedance Z used in the transfer function of the elastic cavity model c 1. The blood viscosity resistance R, vascular compliance C and fingertip characteristic impedance Z of the subject obtained by the calibration module are used. c 1. The characteristic coefficients used in the transfer function of the Tube-Load model are the characteristic coefficients of the Tube-Load model adapted by searching the mapping relationship in the aforementioned characteristic input module. Therefore, in the present invention, in the process of converting the fingertip PPG signal of the subject to be tested into the arterial blood pressure waveform ABP(t) of the radial artery, the model used for the conversion is personalized for the subject to be tested, and has higher accuracy and adaptability.

[0070] The elastic cavity model of the present invention is specifically as follows:

[0071] According to Kirchhoff's voltage law and Kirchhoff's current law, Poiseuille's law Construct an elastic cavity equivalent circuit; transform the time domain circuit model into an s-domain circuit model through a certain Laplace transformation. In this elastic cavity model, the s-domain transfer function of the dynamic pressure-volume relationship of the fingertip pulse wave is expressed according to the preprocessed PPG signal and the potential damping factor in the blood circulation:

[0072]

[0073] Where: s represents the Laplace operator; R is the blood viscosity resistance of the subject to be tested, which can be expressed by the formula Get R value, P a is the mean arterial pressure, Q m is the average volume pulse blood flow. C is the vascular compliance C of the subject to be tested, which can be calculated by the formula Get C value, SV is stroke volume, P s is systolic blood pressure, P d is the diastolic pressure. c 1 is the characteristic impedance of the fingertips of the person being tested, which can be expressed by the formula Get Z c 1 value, Q maxIn the present invention, when the test is actually performed, the blood viscosity resistance R, vascular compliance C and fingertip characteristic impedance Z of the above-mentioned test subject are c 1 is obtained by the calibration module.

[0074] Among them, the input signal of the transfer function G(s) is the preprocessed PPG signal, that is, the overall volume pulse blood flow Q(t) of the microcirculation. The input Q(t) is processed by the transfer function G(s), and the output response signal is the pressure waveform P(t) of the fingertip pulse wave.

[0075] The Tube-Load model of the present invention is as follows:

[0076] The Tube-Load model represents the w-domain transfer function of the pressure pulse wave from the fingertip to the radial artery based on the propagation and reflection phenomena of the pulse wave:

[0077]

[0078] Where w represents the angular frequency, Δt represents the time it takes for the pulse wave to propagate from the radial artery to the fingertip, and T(w) represents the reflection coefficient. Get, A and B are characteristic coefficients.

[0079] In the Tube-Load model, the input signal of the transfer function H(w) is the pressure waveform P(t) of the fingertip pulse wave, the input Q(t) is processed by the transfer function H(w), and the output response signal is the arterial blood pressure waveform ABP(t) of the radial artery.

[0080] In the above Tube-Load model of the present invention, the characteristic coefficients A and B for calculating the reflection coefficient are determined by the characteristic input module according to the personal characteristics input by the subject. Taking the mobile phone APP as an example, the characteristic input module allows the subject to input his personal characteristics in the form of a UI interface, and the personal characteristics include three indicators: age, gender, and the presence or absence of abnormal cardiovascular status.

[0081] In the embodiment of the present invention, in the feature input module, the mapping relationship between the stored personal features and the Tube-Load model feature coefficients is obtained by an adaptive system identification method, and the specific identification method is:

[0082] First, the personal characteristics are divided into multiple categories according to different combinations of three indicators: age, gender, and whether or not there is an abnormal cardiovascular state. For example, age can be divided into multiple age groups, gender can be divided into male / female, and whether or not there is an abnormal cardiovascular state can be divided into abnormal and non-abnormal. Thus, the combination of these different options can be arranged and combined to form multiple categories.

[0083] Then, the forward pressure wave Pf(t) and the reflected pressure wave Pb(t) are used as the input and output of the system identification, respectively, and the reflection coefficient is calculated using the sample data of each category. Regression is performed to determine the values ​​of characteristic coefficients A and B corresponding to each category, thereby establishing a mapping relationship between personal characteristic categories and characteristic coefficients of the Tube-Load model.

[0084] The adaptive system identification method is used to solve the values ​​of A and B in the calculation formula of the reflection coefficient, which avoids the errors caused by physiological changes and solves the problem of model personalization.

[0085] It should be noted that, in the above-mentioned system identification process, before each category determines A and B by regression, it is necessary to collect sample data of the category, and each sample should contain the forward pressure wave Pf(t) and reflected pressure wave Pb(t) data corresponding to an individual of the category. In an embodiment of the present invention, human physiological signals in the mimic database can be used as the above-mentioned samples for system identification. The mimic database is a large public database that records relevant data of patients in the intensive care unit of Beth Israel Deacon Medical Center from 2001 to 2019. It has medical health data and records of more than 40,000 patients, and the recorded demographic characteristics and health data of the patients include gender, age, electrocardiogram signal, heart rate signal, respiratory signal, photoelectric volume pulse wave signal and blood pressure signal. At present, four versions of the mimic database have been developed, and the demographic characteristics and health data of the patients recorded in each version are different. The present invention mainly uses the data of MIMIC-I, and the demographic characteristics and health data of the patients used include age, gender, cardiac output (CO), heart rate (HR), disease type, PPG signal and ABP signal. Through this database, all patient samples can be classified, and then each category can be systematically identified to obtain the A and B values ​​corresponding to the category. The specific solution method based on this data set is:

[0086] S4051: The pressure waveform P(t) of the fingertip pulse wave is obtained by the formula P(t)=Q(t)*g(t), where g(t) represents the result of inverse Laplace transform of the transfer function of the elastic cavity model.

[0087] S4052: By calculating the formula Pf(t) = (P(t) + Q(t) · Z c 2) / 2 and Pb(t)=(P(t)-Q(t)·Z c 2) / 2 solve the forward pressure wave Pf(t) and the reflected pressure wave Pb(t); where Z c 2 represents the characteristic impedance of a purely elastic lossless tube from the radial artery to the fingertip.

[0088] S4053: Pf(t) and Pb(t) are used as the input and output of system identification respectively, and are input into the MATLAB system identification toolbox based on the least squares method principle to process the discrete sampling data of each category.

[0089] S4054: Obtain the values ​​of A and B in the reflection coefficient corresponding to each category.

[0090] Of course, the above-mentioned system identification through the mimic database is only one implementation of the present invention. In practical applications, if a sufficient amount of samples can be sampled, other sample data sets can also be used for classification system identification.

[0091] Based on the above elastic cavity model and Tube-Load model, a hybrid model can be established by combining the two, and its expression is as follows:

[0092] ABP(t)=Q(t)*g(t)*h(t)

[0093] Among them, ABP(t) represents the hybrid model result, g(t) represents the result of inverse Laplace transform of the transfer function of the elastic cavity model, h(t) represents the result of inverse Laplace transform of the transfer function of the Tube-Load model, and the symbol * represents the convolution operation. In practical application, the PPG signal of any subject is preprocessed to form Q(t) and then input into the above hybrid model to obtain the ABP signal based on the individual characteristics of the subject.

[0094] Figure 5 The estimated ABP signal and the measured ABP signal of the present invention are shown, represented by a solid line and a dashed line, respectively. Figure 5 It can be seen that the estimated ABP signal and the measured ABP signal have good correlation and consistency in both amplitude and waveform morphology.

[0095] It should be noted that the hybrid continuous arterial blood pressure waveform detection system in the present invention is essentially a software system running on a hardware system such as a mobile terminal or a host computer. When the detection system runs on a mobile terminal, it can be implemented in the form of an APP, a small program, etc. When the detection system runs on a host computer such as a PC or a server, it can be implemented in the form of a client, a web page, etc. Correspondingly, the signal data that needs to be input in the hardware system, such as the fingertip PPG signal, the reference blood pressure signal, and the cardiac output CO, need to be collected by an external acquisition device. The external acquisition device used to obtain the fingertip PPG signal, the reference blood pressure signal, and the cardiac output CO in the present invention belongs to the prior art and can be implemented by using medical devices corresponding to each indicator. Each medical device can communicate with the hardware system through Bluetooth or other means to send and receive instructions and data to each other.

[0096] In an embodiment of the present invention, the above-mentioned detection system is preferably run on a smart phone, and the APP on the smart phone establishes a connection with the PPG signal acquisition device via Bluetooth. By receiving the blood pressure detection instruction, Bluetooth is turned on to establish a matching connection between the mobile phone APP and the PPG signal acquisition device. When the user to be tested performs blood pressure detection, the server will first establish a server socket during Bluetooth transmission, and then the socket will begin to monitor whether the client has a connection request. The client will also establish a client socket to initiate a connection to the server. If there is no abnormality at this time, the two devices have been paired successfully. Since both the client and the server will hold a Socket at this time, data can be sent and received using the Socket.

[0097] Although the physical condition of the subject remains generally stable over a period of time, it may change over time. Therefore, in an embodiment of the present invention, the calibration procedure in the calibration module needs to be run regularly to keep the parameters in the hybrid model always matching the current state of the subject. In the calibration module, each time the calibration procedure is run, the fingertip PPG signal, reference blood pressure signal and cardiac output CO of the subject need to be collected through an external device, and then the systolic pressure P is obtained from the reference blood pressure signal. s , diastolic blood pressure d and mean arterial pressure P a , the average volume pulse blood flow Q is obtained from the fingertip PPG signal obtained during the calibration phase m and maximum volume pulse blood flow Q max , obtain the stroke volume SV through the cardiac output CO, and then use the formula Obtain the blood viscosity resistance R of the subject to be tested, through the formula Obtain the vascular compliance C of the subject to be tested, through the formula Get the characteristic impedance Z of the fingertip of the person being tested c 1 value, thereby updating the cardiac output CO value, blood viscosity resistance R value, vascular compliance C value and fingertip characteristic impedance Z value of the subject to be tested stored in the system c 1 value.

[0098] In an embodiment of the present invention, the preprocessing of the fingertip PPG signal in the above-mentioned signal processing module can refer to the conventional PPG signal preprocessing process, and mainly needs to eliminate the abnormalities, drifts, interferences and other problems in the original collected PPG signal. The specific preprocessing operation includes: eliminating incomplete signals and abnormal signals, using wavelet transform to process the baseline drift caused by breathing, external light sources and movement factors in the signal, and using bandpass filter to process the high-frequency interference and burrs caused by other electromagnetic in the signal. Wavelet transform has the characteristics of being able to fully highlight certain aspects of the problem, can perform localized analysis of time (space) frequency, and gradually refine the signal (function) at multiple scales through telescopic translation operations, and finally achieve time subdivision at high frequency and frequency subdivision at low frequency, which can automatically adapt to the requirements of time-frequency signal analysis. Therefore, wavelet transform is used to process the baseline drift problem caused by breathing, external light sources and movement. Since 99% of PPG signals are distributed in 0.5-10Hz, in order to remove the interference of other frequency electromagnetics, a bandpass filter is used to process the high-frequency interference and burrs caused by other electromagnetics. For incomplete signals, use the cat() function to splice the captured fragments. This process must ensure the synchronization of the PPG signal and the ABP signal, and capture the complete fragments of the PPG and ABP signals. After the above preprocessing, the processed PPG signal can be calibrated, and the measured PPG signal can be calibrated to the overall volume pulse blood flow Q(t) of the microcirculation according to the patient's cardiac output CO.

[0099] In the present invention, the measured PPG signal is calibrated to the total volume pulse blood flow Q(t) of the microcirculation according to the patient's cardiac output CO, and the specific method can adopt the existing technology. In an embodiment of the present invention, the steps of calibrating the PPG according to the patient's cardiac output CO are as follows:

[0100] (1) For each subject, since CO and HR are known, the stroke volume SV can be obtained;

[0101] (2) Through Q m T = SV (cardiac cycle T is calculated by HR) to obtain the average volume pulse blood flow Q m ;

[0102] (3) Obtain the subject’s P s ,P a and P d The value of

[0103] (4) Through Get the value of R;

[0104] (5) Calculate the waveform feature K1 of the ABP signal And the waveform feature K2 of the PPG signal PPGmax ,PPG min and PPG ave They are the maximum, minimum and average values ​​of PPG respectively;

[0105] (6) Under the linearization assumption, the pulsating components of Q(t) and ABP should be equal, that is, K2(Q s -Q min )=K1(SBP / R-DBP / R), and we get Q s -Q min The value of Q s and Q min They are the maximum volume pulse blood flow and the minimum volume pulse blood flow respectively;

[0106] (7) Finally through Q s -Q min The value and Q m The value of can obtain the total volume pulse blood flow Q(t) defined by CO.

[0107] The hybrid continuous blood pressure detection system composed of the above calibration module, feature input module, signal processing module and signal processing module can essentially be an executable software system. Therefore, based on the same inventive concept, in another embodiment of the present invention, a computer electronic device is provided, which includes a memory and a processor;

[0108] The memory is used to store computer programs;

[0109] The processor is used to implement the above-mentioned hybrid continuous blood pressure detection system when executing the computer program.

[0110] Figure 6 Schematic diagram of a computer electronic device to which the present invention is applicable. Figure 6As shown, a computer electronic device applicable to the present invention mainly consists of five parts, including a controller, an arithmetic unit, a memory, an input device and an output device; the main function of the controller is to schedule programs, data, addresses, coordinate the work of various parts of the computer and control them according to their requirements; the main function of the arithmetic unit is to process data and perform various arithmetic and logical operations; the main function of the memory is to store programs, data and various signals, etc., and provide these information when needed; the input device mainly includes a keyboard, a mouse, an optical disk drive, etc., and its main function includes inputting information such as programs, data, text and control instructions into the computer device; the output device mainly includes a display, a printer, etc., and its The main functions include outputting the intermediate results or final results of the computer device in the form of text, symbols or control instructions; the controller, operator and memory applicable to the present invention are used to schedule the mimic large database, store, process and display the data in the database; the controller, operator and memory applicable to the present invention are used to establish the relationship between input and output of the processed data through several specific computer programs, and the specific computer programs include but are not limited to programs with data processing functions, programs with algorithm development functions, and programs with model optimization functions; the input device applicable to the present invention can be a keyboard, a mouse, etc., and the output device can be a liquid crystal display (Liquid Crystal Display, LCD), an organic light-emitting diode (Organic Light-Emitting Diode, OLED) display or other types of displays such as a touch screen; the input device and the output device are referred to as external devices, which are important components of the computer device. The above-mentioned computer electronic equipment can be implemented using existing PCs, servers, workstations, etc.

[0111] Furthermore, based on the same inventive concept, in another embodiment of the present invention, corresponding to the hybrid continuous blood pressure detection system in the form of the above-mentioned software and the electronic device at the hardware level, the electronic device is coordinated with the remaining external signal acquisition devices, and a hybrid continuous arterial blood pressure waveform detection device based on a single-channel fingertip PPG can be further provided at the hardware level, which includes a paired PPG signal acquisition device, a blood pressure signal acquisition device, a cardiac output acquisition device and an arterial blood pressure detection device.

[0112] The above-mentioned PPG signal acquisition device is used to collect the fingertip PPG signal of the subject after receiving the calibration instruction or blood pressure detection instruction issued by the arterial blood pressure detection device.

[0113] The above-mentioned blood pressure signal acquisition device is used to acquire the reference blood pressure signal of the subject to be measured after receiving the calibration instruction sent by the arterial blood pressure detection device.

[0114] The cardiac output acquisition device is used to acquire the cardiac output CO of the subject after receiving a calibration instruction from the arterial blood pressure detection device.

[0115] The above-mentioned arterial blood pressure detection device is used to send calibration instructions and blood pressure detection instructions to other acquisition devices, and at the same time receive signal data uploaded by each acquisition device; and based on the received data, call the above-mentioned hybrid continuous blood pressure detection system to output the arterial blood pressure waveform ABP(t) of the radial artery of the person to be tested.

[0116] In an embodiment of the present invention, the above-mentioned PPG signal acquisition device, blood pressure signal acquisition device, and cardiac output acquisition device are respectively paired and connected with the arterial blood pressure detection device via Bluetooth modules, signal lines, etc., for two-way transmission of data and instructions, preferably using Bluetooth mode.

[0117] In addition, in an embodiment of the present invention, the above-mentioned PPG signal acquisition device is a wearable device worn on the end of a finger. Figure 7 It is an optional form of PPG signal acquisition device, which is a part of the hybrid continuous arterial blood pressure waveform detection device based on single-channel fingertip PPG provided by the present invention. The PPG signal acquisition device is a non-invasive detection device that detects blood volume changes in living tissue by means of photoelectric means, and the basic principle is Lambert-Beer law. Figure 7 As shown, when the LED emits a light beam of a certain wavelength to illuminate the surface of the fingertip skin, the light beam will be transmitted to the photodetector by transmission or reflection. In this process, due to the absorption and attenuation of the fingertip skin muscle tissue and blood, the light intensity detected by the photodetector will be weakened. The absorption of light by the skin, muscles and tissues remains constant throughout the blood circulation, while the blood volume in the skin changes in a pulsating manner under the contraction and relaxation of the heart. When the heart contracts, the peripheral blood volume is the largest, the light absorption is also the largest, and the detected light intensity is the smallest; while when the heart relaxes, on the contrary, the peripheral blood volume is the smallest, the light absorption is also the smallest, and the detected light intensity is the largest, causing the light intensity detected by the photodetector to change in a pulsating manner, and the light intensity change is converted into an electrical signal, and the change in volume pulse blood flow can be obtained after amplifying and filtering this electrical signal.

[0118] like Figure 8 As shown, the PPG waveform acquisition device includes not only an LED module and a photodetector module, but also other modules; such as a power module, a drive module, a differential amplifier module, a filter module, an ADC module, a Bluetooth transmission module, etc.

[0119] like Figure 8As shown in the figure, the working process of the PPG waveform acquisition device is as follows: first, the detection device is powered by the power module of the 3.3V button lithium-ion battery. When the user to be tested performs blood pressure detection, the fingertip PPG acquisition device is first connected to the mobile phone APP through Bluetooth matching; when the fingertip PPG acquisition device receives the instruction to start detecting the blood pressure signal, it is first initialized, and a light source of a certain wavelength is emitted through the light-emitting diode and the driver module. The LED in the PPG sensor emits green light, and the light source is reflected back to the photodetector after being transmitted and reflected by the various layers of tissue and blood vessels inside the finger; the photodetector converts the optical signal into an electrical signal. Since the signal strength is too weak, it first undergoes a series of differential amplification processes; since other tissues except blood are sensitive to light without significant changes in the human body Absorption is basically unchanged, so the signals we get include DC signals and AC signals. After the signal is processed by the signal bandpass filtering circuit and the AC signal is extracted, an analog PPG signal that reflects the change in blood flow when blood flows in the blood vessels can be obtained; then ADC sampling is performed to obtain a digital PPG signal that reflects the change in blood flow when blood flows in the blood vessels; since ambient light interference is a big problem, especially when modulated light exists, in order to obtain a good signal response, a preprocessing part with ambient light interference elimination and digital filtering parts should be added, and the influence of ambient light interference is reduced through digital signal processing. This is a key function that can very effectively suppress external light interference; the processed signal is used as the input signal, and the PPG signal is displayed above the mobile phone APP interface 2 through Bluetooth data transmission.

[0120] The arterial blood pressure detection device of the present invention can be an electronic device in the form of a mobile device or a host computer, while the above-mentioned hybrid continuous blood pressure detection system exists in the form of software, and provides users with personal characteristic input and arterial blood pressure waveform display functions through a UI interface. Taking a smart phone as an example, the user can enter the personal characteristics of the subject to be tested (age, gender, and whether there is cardiovascular abnormality) on the homepage of the mobile phone APP interface, and view the real-time arterial blood pressure waveform of the subject to be tested in the APP interface. As the control end, the APP design will first plan all the controls to be used, and use the xaml language and constraint layout to complete the design of the APP UI interface; then use the java language to write the code part; first realize the connection and pairing functions with the Bluetooth device, then realize the sending and receiving of data, the formulation of the transmission protocol, and finally realize the display of the dynamic waveform; the specific APP end control process is as follows Fig. 9 As shown: turn on Bluetooth, search for devices, pair devices, and then wait for data to be sent and received; the specific Bluetooth transmission protocol is as follows Fig.10As shown in the Bluetooth Socket communication diagram, the server will first establish a server socket, and then the socket will start to monitor whether the client has a connection request; the client will also establish a client socket to initiate a connection to the server. If there is no abnormality at this time, the two devices have been paired successfully; since both the client and the server have a Socket at this time, data can be sent and received using the Socket. Fig.11 As shown, interface 1 is used to extract the characteristics of the test subjects, and the characteristics of the test subjects are input on the home page of the mobile phone APP interface; the hybrid model based on the characteristics of different groups of people is saved in the database of the mobile phone APP end, and the characteristics of the test subjects input in the mobile phone interface 1 are used to match the specific hybrid model category in the mobile phone interface 2; Fig. 9 The dynamic waveform display part of the APP end control flow chart shown in the figure firstly follows Fig.10 The Bluetooth Socket communication transmission protocol shown displays the collected PPG signal above the mobile phone APP interface 2; then the PPG signal is converted into an ABP signal through a hybrid model and displayed below the mobile phone APP interface 2; and finally, the detection of the individualized arterial blood pressure waveform based on the user to be tested is realized.

Claims

1. A continuous arterial blood pressure waveform detection system based on single-channel fingertip PPG, characterized in that: include: The calibration module is used to obtain the fingertip PPG signal, reference blood pressure signal and cardiac output CO of the subject collected by the external device during the calibration phase, and calibrate the blood viscosity resistance R, vascular compliance C and fingertip characteristic impedance Z of the subject based on the fingertip PPG signal and reference blood pressure signal. c 1; The feature input module is used for the subject to be tested to input his / her personal features, and obtain the feature coefficients of the Tube-Load model adapted to the subject according to the mapping relationship between the pre-stored personal features and the feature coefficients of the Tube-Load model; the personal features include age, gender and the presence or absence of abnormal cardiovascular status; A signal processing module, used for acquiring the fingertip PPG signal of the subject to be tested collected in real time by an external device, and after performing signal preprocessing on the fingertip PPG signal, using the cardiac output CO to calibrate the fingertip PPG signal into the total volume pulse blood flow Q(t) of the subject to be tested; The blood pressure waveform output module is used to input the total volume pulse blood flow Q(t) of the subject to be tested obtained in the signal processing module into a hybrid model composed of an elastic cavity model and a Tube-Load model, and firstly convert the total volume pulse blood flow Q(t) into a pressure waveform P(t) of the fingertip pulse wave by the elastic cavity model, and then convert the pressure waveform P(t) of the fingertip pulse wave into an arterial blood pressure waveform ABP(t) of the radial artery by the Tube-Load model; wherein the blood viscosity resistance R, vascular compliance C and fingertip characteristic impedance Z used in the transfer function of the elastic cavity model are c 1 is obtained by the calibration module, and the characteristic coefficients used in the transfer function of the Tube-Load model are obtained by the characteristic input module.

2. The continuous arterial blood pressure waveform detection system based on single-channel fingertip PPG according to claim 1, characterized in that: The hybrid model is composed of an elastic cavity model and a Tube-Load model. The model input is the total volume pulse blood flow Q(t) of the subject to be tested, and the final output is the arterial blood pressure waveform ABP(t) of the radial artery of the subject to be tested, which is expressed as: ABP(t)=Q(t)*g(t)*h(t) Wherein: the symbol * represents the convolution operation, g(t) is the result of inverse Laplace transform of the s-domain transfer function G(s) of the elastic cavity model, and h(t) is the result of inverse Laplace transform of the w-domain transfer function H(w) of the Tube-Load model; The transfer function G(s) of the elastic cavity model is composed of blood viscosity resistance R, vascular compliance C and fingertip characteristic impedance Z c 1 is determined, s represents the Laplace operator, and the expression is as follows: The transfer function H(w) of the Tube-Load model is determined by the reflection coefficient T(w) and the time Δt for the pulse wave to propagate from the radial artery to the fingertip, and is expressed as follows: The reflection coefficient A and B are the characteristic coefficients, w represents the angular frequency, and j represents the imaginary unit.

3. The continuous arterial blood pressure waveform detection system based on single-channel fingertip PPG according to claim 1, characterized in that: The calibration program in the calibration module is run regularly. Each time the calibration program is run, the fingertip PPG signal, the reference blood pressure signal and the cardiac output CO of the subject to be tested are collected through an external device, and then the systolic pressure P is obtained from the obtained reference blood pressure signal. s , diastolic blood pressure d and mean arterial pressure P a , the average volume pulse blood flow Q is obtained from the fingertip PPG signal obtained during the calibration phase m and maximum volume pulse blood flow Q max , obtain the stroke volume SV through the cardiac output CO, and then use the formula Obtain the blood viscosity resistance R of the subject to be tested, through the formula Obtain the vascular compliance C of the subject to be tested, through the formula Get the characteristic impedance Z of the fingertip of the person being tested c 1 value, thereby updating the cardiac output CO value, blood viscosity resistance R value, vascular compliance C value and fingertip characteristic impedance Z value of the subject to be tested stored in the system c 1 value.

4. The continuous arterial blood pressure waveform detection system based on single-channel fingertip PPG according to claim 1, characterized in that: In the feature input module, the mapping relationship between the stored personal features and the Tube-Load model feature coefficients is obtained by an adaptive system identification method. The specific identification method is: First, the personal characteristics are divided into multiple categories according to different combinations of three indicators: age, gender, and presence or absence of abnormal cardiovascular status; Then, the forward pressure wave Pf(t) and the reflected pressure wave Pb(t) are used as the input and output of the system identification, respectively, and the reflection coefficient is calculated using the sample data of each category. Regression is performed to determine the values ​​of characteristic coefficients A and B corresponding to each category, thereby establishing a mapping relationship between personal characteristic categories and characteristic coefficients of the Tube-Load model.

5. The continuous arterial blood pressure waveform detection system based on single-channel fingertip PPG according to claim 1, characterized in that: In the signal processing module, the preprocessing of the fingertip PPG signal includes: removing incomplete signals and abnormal signals, using wavelet transform to process the baseline drift caused by breathing, external light sources and movement factors in the signal, and using a bandpass filter to process high-frequency interference and glitches caused by other electromagnetic factors in the signal.

6. A computer electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is used to implement the functions corresponding to the continuous arterial blood pressure waveform detection system based on single-channel fingertip PPG as described in any one of claims 1 to 5 when executing the computer program.

7. A continuous arterial blood pressure waveform detection device based on single-channel fingertip PPG, characterized in that: Including paired PPG signal acquisition equipment, blood pressure signal acquisition equipment, cardiac output acquisition equipment and arterial blood pressure detection equipment; The PPG signal acquisition device is used to acquire the fingertip PPG signal of the subject to be tested after receiving the calibration instruction or blood pressure detection instruction issued by the arterial blood pressure detection device; The blood pressure signal acquisition device is used to acquire a reference blood pressure signal of the subject to be measured after receiving a calibration instruction issued by the arterial blood pressure detection device; The cardiac output acquisition device is used to acquire the cardiac output CO of the subject to be tested after receiving the calibration instruction issued by the arterial blood pressure detection device; The arterial blood pressure detection device is used to send calibration instructions and blood pressure detection instructions to other acquisition devices, and at the same time receive signal data uploaded by each acquisition device; and based on the received data, call the continuous arterial blood pressure waveform detection system as described in any one of claims 1 to 5 to output the arterial blood pressure waveform ABP(t) of the radial artery of the subject to be tested.

8. The continuous arterial blood pressure waveform detection device based on single-channel fingertip PPG according to claim 7, characterized in that: The arterial blood pressure detection device is a mobile device or a host computer, and the continuous arterial blood pressure waveform detection system exists in the form of software and provides users with personal feature input and arterial blood pressure waveform display functions through a UI interface.

9. The continuous arterial blood pressure waveform detection device based on single-channel fingertip PPG according to claim 7, characterized in that: The PPG signal acquisition device is a wearable device worn on the end of a finger.

10. The continuous arterial blood pressure waveform detection device based on single-channel fingertip PPG according to claim 7, characterized in that: The PPG signal acquisition device, the blood pressure signal acquisition device, and the cardiac output acquisition device are respectively paired and connected with the arterial blood pressure detection device via Bluetooth modules to perform two-way transmission of data and instructions.

Images