Blood pressure detection method and system based on laser self-mixing interference
By combining laser self-mixing interferometry and deep feedforward neural networks, the problems of accuracy and individual applicability in pulse wave blood pressure measurement have been solved, achieving efficient and accurate blood pressure monitoring suitable for blood pressure detection in multiple scenarios.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING UNIV OF TECH
- Filing Date
- 2023-03-03
- Publication Date
- 2026-05-29
AI Technical Summary
Existing pulse wave blood pressure measurement methods are easily affected by external factors, lack accuracy, and have poor individual applicability. Machine learning methods are far from sufficient in terms of feature learning ability, making it difficult to achieve efficient and accurate blood pressure monitoring.
A laser self-mixing interferometry method is adopted, which uses a laser self-mixing detection system to collect pulse wave signals, obtains feature vectors through signal denoising and reconstruction, and combines them with a deep feedforward neural network to predict blood pressure values. This includes signal denoising, empirical mode decomposition and wavelet transform, and a dual-hidden-layer BP neural network model is constructed.
It achieves high-precision, low-cost continuous blood pressure monitoring, can be applied in multiple scenarios, is suitable for integration into blood pressure monitoring systems, and has the advantages of simple structure and strong applicability.
Smart Images
Figure CN116473528B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of pulse wave blood pressure measurement technology, and in particular to a blood pressure detection method and system based on laser self-mixing interferometry. Background Technology
[0002] Blood pressure, as a commonly used physiological indicator for human health assessment, contains important physiological information, especially for patients in intensive care units who require long-term continuous monitoring to detect abnormalities and adjust treatment plans accordingly. Commonly used blood pressure measurement methods include the Korotkoff sound method, oscillometric method, arterial tone method, and volume compensation method. However, these methods require the measurer to remain still, limiting their applicability and making them structurally complex.
[0003] The human pulse wave is highly correlated with the physiological condition of organs such as the heart and blood vessels, containing a wealth of physiological and pathological information, and can reflect changes in blood pressure to a certain extent. Existing pulse wave-based blood pressure measurement methods have advantages such as simple measurement methods, low requirements for measurement location, small size of measuring devices, and portability. Currently, continuous blood pressure monitoring methods based on pulse waves are mainly divided into two directions: methods based on pulse wave conduction velocity (PWV, or conduction time (PTT)) and methods based on the extraction of pulse wave characteristic parameters.
[0004] Specifically, pulse wave transit time (PWT) refers to the time it takes for the arterial pressure wave to travel from the aortic valve to the peripheral blood vessels during cardiac artery ejection. PWT is a commonly used indicator reflecting arterial elasticity and dilatability. PWT is primarily influenced by blood vessel size and wall elasticity. When blood pressure rises, the blood vessel walls tighten, and blood flow accelerates; when blood pressure falls, the blood vessel walls relax, and blood flow slows. There are two main methods for measuring PWT: the ECG-PPG method, which estimates blood pressure within the same pulse cycle by using the time interval between the R-wave peak of the ECG signal and the peak of the pulse wave signal; and the two-point measurement method based on the same pulse wave conduction tree, which collects pulse wave signals from two different locations on the body and uses the time interval between the two signals as a parameter to calculate blood pressure. The drawbacks of these methods are: they require simultaneous acquisition of two pulse wave signals, making them susceptible to external influences and affecting accuracy. Furthermore, the blood pressure models established using these methods show significant individual variations and lack broad applicability.
[0005] Furthermore, the amplitude and morphology of the pulse wave contain rich information about cardiovascular physiology and pathology, such as blood pressure. The pulse wave parameter measurement method aims to extract feature points from the pulse wave waveform that fully reflect blood pressure. Based on the principles of pulse waves and the theory of arterial elastic cavities, it establishes a correlation between blood pressure and pulse wave feature parameters, thereby building a relevant model to achieve continuous dynamic blood pressure measurement. The drawbacks of the pulse wave feature parameter method are as follows: While current machine learning and big data-based blood pressure estimation methods encompass rich blood pressure information and improve model accuracy to some extent, they still rely heavily on manual extraction of waveform features. Moreover, considering computational costs, these methods only utilize machine learning methods with relatively low time complexity, and their ability to learn features is far from sufficient.
[0006] Therefore, there is an urgent need for an efficient method for measuring blood pressure by pulse wave. Summary of the Invention
[0007] This invention provides a blood pressure detection method and system based on laser self-mixing interferometry, which mainly solves at least one problem in the prior art.
[0008] To achieve the above objectives, the present invention provides a blood pressure detection method based on laser self-mixing interferometry, applied to an electronic device. The method includes: acquiring the pulse wave self-mixing signal of the object to be detected using a laser self-mixing detection system.
[0009] The pulse wave self-mixed signal is subjected to signal denoising and pulse wave curve reconstruction to obtain the feature vector and feature parameters of the pulse wave signal.
[0010] The feature vector of the acquired pulse wave signal is input into the prediction model of human blood pressure value based on deep feedforward neural network; wherein, the prediction model of human blood pressure value based on deep feedforward neural network is obtained by training through the feature parameters.
[0011] The blood pressure value of the subject to be tested is output using the prediction model of human blood pressure value based on deep feedforward neural network.
[0012] Further, preferably, the neural network structure of the prediction model for human blood pressure values based on a deep feedforward neural network includes an input layer, a first hidden layer, a second hidden layer, and an output layer; wherein, the neurons in the input layer are a pre-acquired feature parameter matrix and a feature vector of a pre-acquired pulse wave signal; the output layer is a blood pressure value vector containing systolic blood pressure (SBP) and diastolic blood pressure (DBP); wherein, the feature parameters include pulse wave period, pulse wave propagation time, pulse wave amplitude, and a feature quantity K value based on the change in the area of the pulse wave pattern.
[0013] Furthermore, preferably, after performing signal denoising and pulse wave curve reconstruction on the pulse wave self-mixed signal, the method further includes performing empirical mode decomposition and wavelet transform on the pulse wave self-mixed signal.
[0014] Further, preferably, the laser self-mixing detection system is a vertical-cavity surface-emitting semiconductor laser, which includes a P-type contact layer, a P-type Bragg mirror, an active region, an oxide confinement layer, an N-type Bragg mirror, a substrate, and an N-type contact layer.
[0015] Further, preferably, the method for obtaining the vertical-cavity surface-emitting semiconductor laser includes fabricating an epoxy resin photoresist microlens on a vertical-cavity surface-emitting semiconductor laser chip using flexible wet printing and femtosecond laser two-photon polymerization technology, thereby obtaining the vertical-cavity surface-emitting semiconductor laser.
[0016] A blood pressure detection system based on laser self-mixing interferometry includes a data acquisition module for acquiring the pulse wave self-mixing signal of the object to be detected using a laser self-mixing detection system;
[0017] The feature vector acquisition module is used to perform signal denoising and pulse wave curve reconstruction on the pulse wave self-mixed signal to obtain the feature vector and feature parameters of the pulse wave signal.
[0018] The blood pressure value acquisition module is used to input the feature vector of the acquired pulse wave signal into a prediction model of human blood pressure value based on a deep feedforward neural network; wherein, the prediction model of human blood pressure value based on a deep feedforward neural network is obtained by training through the feature parameters.
[0019] The blood pressure value of the subject to be tested is output using the prediction model of human blood pressure value based on deep feedforward neural network.
[0020] To achieve the above objectives, the present invention also provides an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a program executable by the at least one processor, the program being executed by the at least one processor to enable the at least one processor to perform the laser self-mixing interferometry-based blood pressure detection method as described above.
[0021] In addition, to achieve the above objectives, the present invention also provides a computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps of the above-described blood pressure detection method based on laser self-mixing interference.
[0022] This invention proposes a blood pressure detection method, system, electronic device, and computer-readable storage medium based on laser self-mixing interferometry. The method utilizes a laser self-mixing detection system to acquire the pulse wave self-mixing signal of the object to be detected; performs signal denoising and pulse wave curve reconstruction on the pulse wave self-mixing signal to obtain the feature vector and feature parameters of the pulse wave signal; inputs the acquired feature vector of the pulse wave signal into a prediction model of human blood pressure value based on a deep feedforward neural network; wherein the prediction model of human blood pressure value based on the deep feedforward neural network is obtained through training with the feature parameters; and outputs the blood pressure value of the object to be detected using the prediction model of human blood pressure value based on the deep feedforward neural network. The beneficial effects are as follows: This invention discloses an integrated self-mixing interferometry detection system for blood pressure value detection. This system obtains changes in blood flow velocity, vessel wall displacement, blood absorption of detection light, and transmission time by measuring the pulse wave characteristics of radial artery microvessels, thereby achieving long-term, non-invasive, continuous blood pressure measurement. The laser self-mixing interferometry technology of the present invention can simultaneously detect multiple characteristic parameters of the pulse wave; it has the advantages of low cost, high precision, simple structure and wide range of applicable scenarios, and is very suitable for integration into blood pressure monitoring systems. Attached Figure Description
[0023] Figure 1 This is a flowchart of a preferred embodiment of the blood pressure detection method based on laser self-mixing interferometry of the present invention;
[0024] Figure 2 This is a schematic diagram of the structure of the vertical cavity surface-emitting semiconductor laser of the present invention;
[0025] Figure 3 This is a schematic diagram of the structure of the human blood pressure prediction model based on a deep feedforward neural network according to the present invention.
[0026] Figure 4 This is a schematic diagram of the logical structure of the blood pressure detection system based on laser self-mixing interference of the present invention;
[0027] Figure 5 This is a schematic diagram of a preferred embodiment of the electronic device of the present invention;
[0028] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation
[0029] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
[0030] Definitions:
[0031] The laser self-mixing interference effect refers to the phenomenon where the output light of a laser is partially reflected or scattered by an external object and then fed back into the laser's resonant cavity. This feedback light, carrying information about the external object, interferes with the light inside the cavity, modulating the laser's output characteristics, thereby enabling the measurement of the physical quantities of the target object.
[0032] The pulse wave is formed by the heartbeat propagating outwards along the arteries and blood flow.
[0033] VCSEL, or Vertical Cavity Surface Emitting Laser, is a novel type of semiconductor laser. It emits laser light perpendicular to the substrate surface and is based on gallium arsenide semiconductor material. VCSELs primarily have three structures: 45° mirror type, grating-coupled type, and vertical cavity type. The main structure of a VCSEL consists of two parts: the central active region, which includes both bulk heterojunction and quantum well structures; and its lateral structures, which can be divided into gain-guided and ring-buried heterostructures.
[0034] Wavelet transform (WT) is a method for time-frequency analysis and processing of signals. It inherits and develops the localization concept of the short-time Fourier transform while overcoming the drawbacks of window size not changing with frequency. WT provides a frequency-varying "time-frequency" window, making it an ideal tool for time-frequency analysis and processing. Its main characteristics are its ability to fully highlight certain features of a problem through transformation, its capacity for localized analysis of time (space) and frequency, and its ability to progressively refine the signal (function) at multiple scales through scaling and translation operations. This ultimately achieves time subdivision at high frequencies and frequency subdivision at low frequencies, automatically adapting to the requirements of time-frequency signal analysis. This allows for focusing on arbitrary details of the signal, solving the difficulties of the Fourier transform.
[0035] Empirical Mode Decomposition (EMD) assumes that any signal is composed of different intrinsic mode functions (IMFs). IMFs are a class of signals that satisfy the physical interpretation of a single-component signal, having only a single frequency component at any given time, representing the signal's inherent characteristic vibrational form. IMF components must satisfy two conditions: first, the number of extrema and zero-crossings must be the same or differ by at most one; second, at any given time, the mean of the upper and lower envelopes formed by the extrema is zero. The essence of the EMD method is to obtain the intrinsic vibrational modes of a signal through characteristic time scales, and then continuously "simplify" the signal using these intrinsic vibrational modes.
[0036] This invention discloses an integrated self-mixing interferometry detection system for blood pressure detection. This system measures the pulse wave characteristics of radial artery microvessels to obtain changes in blood flow velocity, vessel wall displacement, blood absorption of probe light, and transmission time, thereby achieving long-term, non-invasive, continuous blood pressure measurement. The laser self-mixing interferometry technology of this invention can simultaneously detect multiple characteristic parameters of the pulse wave; it has advantages such as low cost, high accuracy, simple structure, and wide applicability, making it highly suitable for integration into blood pressure monitoring systems.
[0037] This invention provides a blood pressure detection method based on laser self-mixing interference. Figure 1 The flowchart of a preferred embodiment of the blood pressure detection method based on laser self-mixing interferometry of the present invention is shown. (Refer to...) Figure 1 As shown, the method can be performed by a device, which can be implemented by software and / or hardware.
[0038] It should be noted that the blood pressure detection method based on laser self-mixing interferometry of the present invention specifically includes steps S110-S140.
[0039] S110. Use a laser self-mixing detection system to collect the pulse wave self-mixing signal of the object to be detected.
[0040] Specifically, it collects changes in the blood flow velocity, blood vessel wall displacement, blood absorption of the probe light, and transmission time of the object to be tested.
[0041] Figure 2 This is a schematic diagram of the structure of the vertical cavity surface-emitting semiconductor laser of the present invention; as shown. Figure 2 As shown, the laser self-mixing detection system is a vertical-cavity surface-emitting semiconductor laser (VCSEL). The VCSEL includes a P-type contact layer 1, a P-type Bragg mirror 2, an active region 3, an oxide confinement layer 4, an N-type Bragg mirror 5, a substrate 6, and an N-type contact layer 7. In terms of light source, VCSELs possess unique advantages such as small size, low power consumption, ease of array formation, and high beam quality, making them ideal for laser self-mixing detection systems. Furthermore, the junction voltage of the VCSEL chip can be used as a new self-mixing signal source to replace the embedded photodiode current signal in traditional semiconductor laser self-mixing detectors, further simplifying the blood pressure detection system based on laser self-mixing interferometry.
[0042] In one specific embodiment, the method for obtaining a vertical-cavity surface-emitting diode (VCSEL) laser includes fabricating an epoxy resin photoresist microlens on a VCSEL chip using flexible wet printing and femtosecond laser two-photon polymerization (FPSP) technology, thereby obtaining the VCSEL. Specifically, in the optical system, an epoxy resin photoresist microlens is directly fabricated on the VCSEL chip using existing flexible wet printing technology combined with FPSP technology to replace the glass lens in the prior art, effectively focusing the beam within a distance of 1-2 mm and improving the self-mixing signal intensity. Furthermore, a dry cell battery is used instead of the original DC power supply to power the detection system, thereby achieving miniaturization of the blood pressure detection system based on laser self-mixing interferometry.
[0043] S120. Perform signal denoising and pulse wave curve reconstruction on the pulse wave self-mixed signal to obtain the feature vector and feature parameters of the pulse wave signal.
[0044] After performing signal denoising and pulse wave curve reconstruction on the pulse wave self-mixed signal, the method further includes performing empirical mode decomposition and wavelet transform on the pulse wave self-mixed signal.
[0045] In the specific implementation process, the characteristic parameters include pulse wave period, pulse wave propagation time, pulse wave amplitude, and the characteristic quantity K value based on the change in the area of the pulse wave pattern. Specifically, the pulse wave characteristic parameters can include pulse wave time-domain parameters, pulse wave frequency-domain parameters, and pulse wave statistical parameters. Among them, the pulse wave time-domain parameters are parameters obtained by comprehensively analyzing the position of characteristic points in the pulse wave sampling sequence on the time series, i.e., time point, amplitude, and the area enclosed by each point. The characteristic points here include the peak, trough, inflection point, and dichroism point of the pulse wave. The pulse wave frequency-domain parameters are parameters obtained by comprehensively analyzing the position of characteristic points in the pulse wave spectrum on the frequency domain sequence, i.e., frequency point, amplitude, and the area enclosed by each point. The characteristic points here include the maximum peak point, each harmonic peak point, and specific frequency points. The pulse wave statistical parameters are parameters obtained by statistically constructing the pulse wave time-domain parameters and pulse wave frequency-domain parameters of multiple pulse wave signals. In specific implementation, pulse wave characteristic parameters may include, but are not limited to, pulse wave period, pulse wave propagation time, pulse wave amplitude, and the characteristic quantity K value based on the change in pulse wave area. In summary, this invention obtains a self-mixed pulse wave signal by measuring the radial artery using self-mixing interferometry, reconstructs the pulse wave curve using noise reduction techniques such as EMD and wavelet transform, and then extracts the pulse wave characteristic parameters as input layer neurons of a BP neural network. After passing through a double-hidden-layer neural network blood pressure model, it outputs systolic and diastolic blood pressure, thereby achieving blood pressure measurement.
[0046] The signal is preprocessed to reduce noise, and the relevant features of the vibration signal after noise reduction are extracted to characterize different blood pressure values, thereby forming feature samples.
[0047] S130. Input the feature vector of the acquired pulse wave signal into the prediction model of human blood pressure value based on deep feedforward neural network; wherein, the prediction model of human blood pressure value based on deep feedforward neural network is obtained by training through the feature parameters.
[0048] The neural network structure of the human blood pressure prediction model based on a deep feedforward neural network includes an input layer, a first hidden layer, a second hidden layer, and an output layer. The input layer neurons are pre-acquired feature parameter matrices and feature vectors of pre-acquired pulse wave signals. The output layer contains a blood pressure value vector including systolic blood pressure (SBP) and diastolic blood pressure (DBP). Specifically, in terms of data processing, feature points of the pulse wave signal are extracted from the measured laser self-mixing vibration signal using methods such as empirical mode decomposition and wavelet transform. Then, a prediction model for human blood pressure is constructed using a deep feedforward neural network. The neural network structure of the human blood pressure prediction model based on a deep feedforward neural network is as follows: Figure 3 As shown, the feature parameter matrix uses the blood pressure-related feature vector X extracted in previous work as the input layer neuron, and the output layer is the vector Y containing the systolic blood pressure SBP and diastolic blood pressure DBP.
[0049] In the specific implementation process, a dual-hidden-layer BP neural network structure is constructed. The optimal number of nodes in the first hidden layer and the optimal number of nodes in the second hidden layer are determined using empirical formulas and fuzzy inference, respectively. The variable-node dual-hidden-layer BP neural network is trained by dividing the extracted feature samples into a training set and a test set. The training set is used to train the variable-node dual-hidden-layer BP neural network to obtain its relevant parameters. The variable-node dual-hidden-layer BP neural network is then tested by inputting the test set into the trained variable-node dual-hidden-layer BP neural network, which outputs the blood pressure status of the target.
[0050] S140. Utilize the prediction model of human blood pressure values based on the deep feedforward neural network to output the blood pressure value of the object to be detected.
[0051] This invention discloses a blood pressure detection method based on laser self-mixing interferometry. The method utilizes a laser self-mixing detection system to acquire the self-mixed pulse wave signal of the target object. Noise reduction and pulse wave curve reconstruction are performed on the self-mixed pulse wave signal to obtain its feature vector and feature parameters. The acquired feature vector is input into a prediction model for human blood pressure values based on a deep feedforward neural network. This prediction model is obtained through training with the feature parameters. The blood pressure value of the target object is output using the prediction model. The beneficial effects are as follows: This invention discloses an integrated self-mixing interferometry detection system for blood pressure detection. This system measures the pulse wave characteristics of radial artery microvessels to obtain changes in blood flow velocity, vessel wall displacement, blood absorption of probe light, and transmission time, thereby achieving long-term, non-intrusive, continuous blood pressure measurement. The laser self-mixing interferometry technology of this invention can simultaneously detect multiple feature parameters of the pulse wave; it has advantages such as low cost, high accuracy, simple structure, and wide applicability, making it highly suitable for integration into blood pressure monitoring systems.
[0052] Figure 4 This is a schematic diagram of the logical structure of the blood pressure detection system based on laser self-mixing interferometry of the present invention; see reference. Figure 4 As shown, to achieve the above objectives, the present invention provides a blood pressure detection system 400 based on laser self-mixing interferometry, including a data acquisition module 410, a feature vector acquisition module 420, and a blood pressure value acquisition module 430. Wherein,
[0053] Includes a data acquisition module 410, used to acquire the pulse wave self-mixing signal of the object to be detected using a laser self-mixing detection system;
[0054] The feature vector acquisition module 420 is used to perform signal denoising and pulse wave curve reconstruction on the pulse wave self-mixed signal to obtain the feature vector and feature parameters of the pulse wave signal.
[0055] The blood pressure acquisition module 430 is used to input the feature vector of the acquired pulse wave signal into a prediction model of human blood pressure value based on a deep feedforward neural network; wherein, the prediction model of human blood pressure value based on a deep feedforward neural network is obtained by training through the feature parameters; and the blood pressure value of the subject to be detected is output using the prediction model of human blood pressure value based on a deep feedforward neural network.
[0056] In summary, this invention is a blood pressure detection system based on laser self-mixing interferometry. It discloses an integrated self-mixing interferometry detection system for blood pressure measurement. This system obtains changes in blood flow velocity, vessel wall displacement, blood absorption of the probe light, and transmission time by measuring the pulse wave characteristics of the radial artery microvessels, thereby achieving long-term, non-invasive, continuous blood pressure measurement. The laser self-mixing interferometry technology of this invention can simultaneously detect multiple characteristic parameters of the pulse wave; it has advantages such as low cost, high accuracy, simple structure, and wide applicability, making it highly suitable for integration into blood pressure monitoring systems.
[0057] This invention provides a blood pressure detection method based on laser self-mixing interference, which is applied to an electronic device 5.
[0058] Figure 5 The application environment of a preferred embodiment of the blood pressure detection method based on laser self-mixing interferometry according to the present invention is shown.
[0059] Reference Figure 5 As shown, in this embodiment, the electronic device 5 can be a terminal device with computing capabilities, such as a server, smartphone, tablet computer, portable computer, or desktop computer.
[0060] The electronic device 5 includes: a processor 52, a memory 51, a communication bus 53, and a network interface 55.
[0061] The memory 51 includes at least one type of readable storage medium. The at least one type of readable storage medium may be a non-volatile storage medium such as flash memory, hard disk, multimedia card, card-type memory 51, etc. In some embodiments, the readable storage medium may be an internal storage unit of the electronic device 5, such as the hard disk of the electronic device 5. In other embodiments, the readable storage medium may also be an external memory 51 of the electronic device 5, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 5.
[0062] In this embodiment, the readable storage medium of the memory 51 is typically used to store a laser self-mixing interferometry-based blood pressure detection program 50, etc., installed on the electronic device 5. The memory 51 can also be used to temporarily store data that has been output or will be output.
[0063] In some embodiments, processor 52 may be a central processing unit (CPU), microprocessor or other data processing chip, used to run program code stored in memory 51 or process data, such as executing a blood pressure detection program 50 based on laser self-mixing interferometry.
[0064] The communication bus 53 is used to enable communication between these components.
[0065] The network interface 54 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface), which is typically used to establish communication connections between the electronic device 5 and other electronic devices.
[0066] Figure 5 Only electronic device 5 with components 51-54 is shown; however, it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.
[0067] Optionally, the electronic device 5 may also include a user interface, which may include an input unit such as a keyboard, a voice input device such as a microphone or other device with voice recognition function, a voice output device such as a speaker or headphones, etc. Optionally, the user interface may also include a standard wired interface or a wireless interface.
[0068] Optionally, the electronic device 5 may also include a display, which may also be referred to as a display screen or display unit. In some embodiments, it may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an organic light-emitting diode (OLED) touchscreen, etc. The display is used to display information processed in the electronic device 5 and to display a visual user interface.
[0069] Optionally, the electronic device 5 may also include radio frequency (RF) circuits, sensors, audio circuits, etc., which will not be described in detail here.
[0070] exist Figure 5In the illustrated device embodiment, the memory 51, which serves as a computer storage medium, may include an operating system and a blood pressure detection program 50 based on laser self-mixing interferometry. When the processor 52 executes the blood pressure detection program 50 based on laser self-mixing interferometry stored in the memory 51, it performs the following steps: acquiring the pulse wave self-mixing signal of the object to be detected using a laser self-mixing detection system; performing signal denoising and pulse wave curve reconstruction on the pulse wave self-mixing signal to obtain the feature vector and feature parameters of the pulse wave signal; inputting the acquired feature vector of the pulse wave signal into a prediction model of human blood pressure value based on a deep feedforward neural network; wherein, the prediction model of human blood pressure value based on a deep feedforward neural network is obtained by training with the feature parameters; and outputting the blood pressure value of the object to be detected using the prediction model of human blood pressure value based on a deep feedforward neural network.
[0071] In other embodiments, the blood pressure detection program 50 based on laser self-mixing interferometry can be further divided into one or more modules, which are stored in memory 51 and executed by processor 52 to complete the present invention. The module referred to in this invention is a series of computer program segments capable of performing specific functions. The blood pressure detection program 50 based on laser self-mixing interferometry can be divided into the following steps: acquiring the pulse wave self-mixing signal of the object to be detected using a laser self-mixing detection system; performing signal denoising and pulse wave curve reconstruction on the pulse wave self-mixing signal to obtain the feature vector and feature parameters of the pulse wave signal; inputting the acquired feature vector of the pulse wave signal into a prediction model of human blood pressure value based on a deep feedforward neural network; wherein the prediction model of human blood pressure value based on a deep feedforward neural network is obtained through training with the feature parameters; and outputting the blood pressure value of the object to be detected using the prediction model of human blood pressure value based on a deep feedforward neural network.
[0072] Furthermore, the present invention also proposes a computer-readable storage medium, which mainly includes a data storage area and a program storage area. The data storage area can store data created based on the use of blockchain nodes, etc., and the program storage area can store an operating system and an application program required for at least one function. The computer-readable storage medium includes a blood pressure detection program based on laser self-mixing interferometry. When the blood pressure detection program based on laser self-mixing interferometry is executed by a processor, it implements the operation of a blood pressure detection method based on laser self-mixing interferometry.
[0073] The specific implementation of the computer-readable storage medium of the present invention is largely the same as the specific implementation of the blood pressure detection method, system and electronic device based on laser self-mixing interference described above, and will not be repeated here.
[0074] In summary, this invention relates to a method, system, electronic device, and computer-readable storage medium for blood pressure detection based on laser self-mixing interferometry. It utilizes a laser self-mixing detection system to acquire the pulse wave self-mixing interferometry signal of the object to be detected; performs signal denoising and pulse wave curve reconstruction on the pulse wave self-mixing interferometry signal to obtain the feature vector and feature parameters of the pulse wave signal; inputs the acquired feature vector of the pulse wave signal into a prediction model of human blood pressure value based on a deep feedforward neural network; wherein the prediction model of human blood pressure value based on the deep feedforward neural network is obtained through training with the feature parameters; and outputs the blood pressure value of the object to be detected using the prediction model of human blood pressure value based on the deep feedforward neural network. The beneficial effects are as follows: This invention discloses an integrated self-mixing interferometry detection system for blood pressure value detection. This system obtains changes in blood flow velocity, vessel wall displacement, blood absorption of probe light, and transmission time by measuring the pulse wave characteristics of radial artery microvessels, thereby achieving long-term, non-invasive, continuous blood pressure measurement. The laser self-mixing interferometry technology of the present invention can simultaneously detect multiple characteristic parameters of the pulse wave; it has the advantages of low cost, high precision, simple structure and wide range of applicable scenarios, and is very suitable for integration into blood pressure monitoring systems.
[0075] The blockchain referred to in this invention is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.
[0076] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.
[0077] The sequence numbers of the above embodiments of the present invention are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases, the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several programs to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.
[0078] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.
Claims
1. A blood pressure detection method based on laser self-mixing interferometry, applied to electronic devices, characterized in that, This method is used to achieve long-term, unobtrusive, continuous blood pressure measurement; the methods include: A laser self-mixing detection system was used to acquire pulse wave self-mixing signals at the radial artery microvessels of the object under test. The pulse wave self-mixed signal is subjected to signal denoising and pulse wave curve reconstruction to obtain the feature vector and feature parameters of the pulse wave signal. The feature vector of the acquired pulse wave signal is input into the prediction model of human blood pressure value based on deep feedforward neural network; wherein, the prediction model of human blood pressure value based on deep feedforward neural network is obtained by training through the feature parameters. The blood pressure prediction model based on a deep feedforward neural network outputs the blood pressure value of the subject to be tested. The neural network structure of the prediction model includes an input layer, a first hidden layer, a second hidden layer, and an output layer. The input layer neurons are a pre-acquired feature parameter matrix and a feature vector of a pre-acquired pulse wave signal. The output layer contains a blood pressure value vector including systolic blood pressure (SBP) and diastolic blood pressure (DBP). The feature parameters include pulse wave period, pulse wave propagation time, pulse wave amplitude, and a feature quantity K value based on the change in pulse wave area. The laser self-mixing detection system is a vertical-cavity surface-emitting semiconductor laser, which includes a P-type contact layer, a P-type Bragg mirror, an active region, an oxide confinement layer, an N-type Bragg mirror, a substrate, and an N-type contact layer.
2. The blood pressure detection method based on laser self-mixing interferometry according to claim 1, characterized in that, After performing signal denoising and pulse wave curve reconstruction on the pulse wave self-mixed signal, the method further includes performing empirical mode decomposition and wavelet transform on the pulse wave self-mixed signal.
3. The blood pressure detection method based on laser self-mixing interferometry according to claim 1, characterized in that, The method for obtaining the vertical-cavity surface-emitting semiconductor laser includes: An epoxy resin photoresist microlens was fabricated on a vertical-cavity surface-emitting semiconductor laser chip using flexible wet printing and femtosecond laser two-photon polymerization technology, thus obtaining a vertical-cavity surface-emitting semiconductor laser.
4. A blood pressure detection system based on laser self-mixing interferometry, characterized in that, This method is used to achieve long-term, non-invasive, continuous blood pressure measurement using the laser self-mixing interference-based blood pressure detection method as described in claim 1. include The acquisition module is used to acquire pulse wave self-mixing signals at the radial artery microvessels of the object to be detected using a laser self-mixing detection system; The feature vector acquisition module is used to perform signal denoising and pulse wave curve reconstruction on the pulse wave self-mixed signal to obtain the feature vector and feature parameters of the pulse wave signal. The blood pressure value acquisition module is used to input the feature vector of the acquired pulse wave signal into a prediction model of human blood pressure value based on a deep feedforward neural network; wherein, the prediction model of human blood pressure value based on a deep feedforward neural network is obtained by training through the feature parameters. The blood pressure value of the subject to be tested is output using the prediction model of human blood pressure value based on deep feedforward neural network.
5. An electronic device, characterized in that, The electronic device includes: at least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a program that is executed by the at least one processor to enable the at least one processor to perform the laser self-mixing interferometry-based blood pressure detection method as described in any one of claims 1 to 3.
6. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the blood pressure detection method based on laser self-mixing interference as described in any one of claims 1 to 3.