A three-wavelength blood oxygen saturation monitoring method based on wearable devices

By using three-wavelength spectrum and dynamic adaptive models on wearable devices, the inaccuracy and discomfort problems of existing blood oxygen saturation measurement methods are solved, and accurate blood oxygen saturation monitoring is achieved that adapts to different environments and user status.

CN113274008BActive Publication Date: 2024-11-08BEIJING XUEYANG TECH CO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202110428234.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-04-21
Publication Date
2024-11-08
Estimated Expiration
2041-04-21

AI Technical Summary

Technical Problem

The existing blood oxygen saturation measurement methods are troublesome, uncomfortable, and cannot achieve continuous monitoring. They are greatly affected by the environment and user status, resulting in inaccurate measurements.

Method used

Using a three-wavelength blood oxygen saturation monitoring method based on wearable devices, the blood oxygen saturation is calculated based on a dynamic adaptive model by obtaining green light, red light and infrared light pulse wave signals, and using Chebischev filter and green light filter The mechanism performs signal processing to adapt to different environments and states.

Benefits of technology

It realizes more accurate and comfortable blood oxygen saturation monitoring, adapts to different environments and user states, and improves the robustness and adaptability of measurement.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113274008B_ABST
    Figure CN113274008B_ABST
Patent Text Reader

Abstract

The present invention provides a three-wavelength blood oxygen saturation monitoring method based on a wearable device, comprising: obtaining a pulse wave signal of an object to be detected; wherein the pulse wave signal comprises a green light pulse wave signal, a red light pulse wave signal and an infrared light pulse wave signal; based on the environment of the object to be detected, obtaining a state filtering range of the green light pulse wave signal, and generating a green light filtering mechanism according to the state filtering range; searching for red light feature points according to the red light pulse wave signal, and determining a first signal component based on the red light feature points; searching for infrared light feature points according to the infrared light pulse wave signal, and determining a second signal component based on the infrared light feature points; filtering the first signal component and the second signal component through the green light filtering mechanism, and establishing a dynamic adaptive model after the processing; and determining the blood oxygen saturation of the object to be detected based on the dynamic adaptive model.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of blood oxygen saturation monitoring, and in particular to a three-wavelength blood oxygen saturation monitoring method based on a wearable device. Background Art

[0002] The blood sample saturation reflects the percentage of hemoglobin bound to oxygen in the blood, that is, monitoring blood oxygen saturation is an estimate of hemoglobin bound to lung oxygen, and is an important means and method for clinical detection of hypoxia.

[0003] The traditional method of measuring blood oxygen saturation is to use a blood gas analyzer to collect blood from the human body, and then perform electrochemical analysis on the blood sample to measure the blood oxygen partial pressure and calculate the blood oxygen saturation. This method is not only cumbersome, but also requires fresh blood from the patient or user, and is a passive physical test, and cannot continuously monitor patients or users in poor physical condition. The existing wrist technology uses a finger-cuff photoelectric sensor, using a wavelength of 660nm red light and 940nm near-infrared light as the incident light source, using the finger as a transparent container for hemoglobin, and measuring the light transmission intensity through the tissue bed to calculate the hemoglobin concentration and blood oxygen saturation.

[0004] The existing technology uses a single red light and infrared light signal to calculate the wavelength coefficient, and then uses an empirical formula to solve the blood oxygen saturation. It is affected by the environment and the user's physical state, has strong interference, is uncomfortable to wear, has poor robustness, has no adaptability, and uses a single wavelength coefficient. For these reasons, the existing technology cannot accurately measure the blood oxygen saturation. Summary of the invention

[0005] The present invention provides a three-wavelength blood oxygen saturation monitoring method based on a wearable device to solve the problems arising from the above-mentioned background technology.

[0006] The present invention provides a three-wavelength blood oxygen saturation monitoring method based on a wearable device, which is characterized by comprising:

[0007] Acquire the pulse wave signal of the object to be detected; wherein,

[0008] The pulse wave signal includes a green light pulse wave signal, a red light pulse wave signal and an infrared light pulse wave signal;

[0009] Based on the environment of the object to be detected, a state filtering range of the green light pulse wave signal is obtained, and a green light filtering mechanism is generated according to the state filtering range;

[0010] According to the red light pulse wave signal, searching for red light characteristic points, and determining a first signal component based on the red light characteristic points;

[0011] According to the infrared light pulse wave signal, searching for infrared light characteristic points, and determining a second signal component based on the infrared light characteristic points;

[0012] The first signal component and the second signal component are filtered by a green light filtering mechanism, and a dynamic adaptive model is established after the processing;

[0013] The blood oxygen saturation of the object to be detected is calculated based on the dynamic adaptive model.

[0014] Preferably, before acquiring the pulse wave signal of the object to be detected, the method further includes:

[0015] Based on the preset sensing device in the wearable device, determine whether the object to be detected is in a stationary state:

[0016] When it is detected that the object to be detected is in a stationary state, based on the preset wearable device, a collection instruction is automatically issued to collect the red light pulse wave signal and infrared light pulse signal of the object to be detected;

[0017] When it is detected that the object to be detected is not in a static state, based on the preset wearable device, a collection instruction is automatically issued to collect the green light pulse wave signal of the object to be detected.

[0018] Preferably, the step of obtaining a pulse wave signal of the object to be detected when the object is in a stationary state includes:

[0019] Based on the preset wearable device, light waves are emitted alternately to the object to be detected in a stationary state at regular intervals; wherein,

[0020] The light wave rays include green light rays, red light rays and infrared light rays;

[0021] Acquiring the light intensity of the red light and the infrared light, and determining the arterial pulsation optical path of the object to be detected according to the light intensity;

[0022] According to the arterial pulsation optical path, the pulse intensity of the object to be detected is recorded in real time;

[0023] A pulse wave signal of the object to be detected is determined according to the pulse intensity.

[0024] Preferably, the state filtering range of the green light pulse wave signal is obtained based on the environment of the object to be detected, and a green light filtering mechanism is generated according to the state filtering range, including:

[0025] acquiring a green light pulse wave signal of the object to be detected at a regular interval, and determining a characteristic frequency of the green light pulse wave signal according to the green light pulse wave signal;

[0026] According to the characteristic frequency, the signal influencing parameter of the pulse wave signal is obtained; wherein,

[0027] The signal influencing parameters include the fluctuation coefficient, cut-off frequency and low pass resistance order of the pulse wave signal;

[0028] According to the signal impact parameters, the maximum attenuation data is calculated; wherein,

[0029] The loss data is loss data under the influence of the wearable device's own parameters;

[0030] Based on the environment of the object to be detected, the maximum attenuation data is transmitted to a preset Chebyshev filter, and a green light filtering mechanism is established.

[0031] Preferably, based on the environment of the object to be detected, the maximum attenuation data is transmitted to a preset Chebyshev filter, and a green light filtering mechanism is established, including:

[0032] Based on the environment of the object to be detected, the green light filtering frequency is collected through a preset Chebyshev filter; wherein,

[0033] The filtering frequency is the noise frequency and abnormal signal of the object to be detected in the environment;

[0034] Calculating the pulse wave characteristic matrix of the object to be detected in a non-stationary state;

[0035] Acquire and transmit the maximum attenuation data to the Chebyshev filter, and filter the pulse wave characteristic matrix through the green light filtering frequency to determine the target filtering matrix;

[0036] A green light filtering mechanism is established according to the target filtering matrix.

[0037] Preferably, the step of calculating the pulse wave characteristic matrix of the object to be detected in a non-stationary state includes:

[0038] When the object to be detected is in a non-stationary state, a green light pulse wave signal of the object to be detected is collected;

[0039] Determining green light pulse wave characteristic points according to the green light pulse wave signal;

[0040] Fitting and calculating the pulse wave characteristic points, and determining the characteristic coefficients of the green light pulse characteristic points;

[0041] The pulse wave characteristic points are matrixed according to the characteristic coefficients to determine a pulse wave characteristic matrix.

[0042] Preferably, establishing a green light filtering mechanism according to the target filtering matrix includes:

[0043] Step 1: Get the preset ideal filter matrix

[0044]

[0045] Wherein, Density represents the ideal filter matrix, n×n represents the ideal filter matrix is ​​a matrix with n rows and n columns, and density represents the pulse wave characteristic data under the ideal filtering condition in the ideal filter matrix;

[0046] Step 2: Get the target filter matrix

[0047]

[0048] Wherein, target represents the target filter matrix, n×n represents the target filter matrix is ​​a matrix with n rows and n columns, and target represents the pulse wave characteristic data under the target filtering condition in the target filter matrix;

[0049] Step 3: Determine a transform coefficient matrix according to the target filter matrix and the ideal filter matrix;

[0050]

[0051] Among them, W n×n represents the transformation coefficient matrix, E n×n represents the identity matrix;

[0052] Step 4: Establish a green light filter spectrum matrix through the transformation coefficient matrix;

[0053]

[0054] Where r represents the green light filter spectrum matrix;

[0055] Step 5: Determine a green light filtering mechanism according to the green light filtering spectrum matrix.

[0056] Preferably, searching for red light feature points according to the red light pulse wave signal, and determining the first signal component based on the red light feature points, comprises:

[0057] Calculating the red light pulse wave intensity according to the red light pulse wave signal;

[0058] Sorting the intensity of the red light pulse wave and determining a first intensity peak;

[0059] Collecting and reconstructing red light feature points through the first intensity peak;

[0060] According to the red light feature points, the red light DC component and the red light AC component are calculated; wherein,

[0061] The red light AC component is a change signal of arterial pulse obtained based on photoplethysmography under the action of red light;

[0062] A first signal component is determined according to the red light DC component and the red light AC component.

[0063] Preferably, searching for infrared light feature points according to the infrared light pulse wave signal, and determining the second signal component based on the infrared light feature points, comprises:

[0064] Calculating the infrared light pulse wave intensity according to the infrared light pulse wave signal;

[0065] Sorting the intensity of the infrared light pulse wave and determining a second intensity peak;

[0066] Collecting and reconstructing infrared light feature points through the second intensity peak;

[0067] According to the red light external feature points, the infrared light DC component and the infrared light AC component are calculated; wherein,

[0068] The infrared light AC component is a change signal of the arterial pulse obtained based on photoplethysmography under the action of the infrared light;

[0069] A second signal component is determined according to the infrared light DC component and the infrared light AC component.

[0070] Preferably, the first signal component and the second signal component are filtered through a green light filtering mechanism, and a dynamic adaptive model is established after the processing, the first signal component and the second signal component are filtered through a green light filtering mechanism, and a dynamic adaptive model is established after the processing, including:

[0071] Step S1: Collect the first signal component Δsig1 (f d1 ,f s1 ) and the second signal component Δsig2(f d2 ,f s2 );

[0072] Where Δsig1 represents the first signal component, f d1 represents the first DC component of the first signal component, f s1 represents the first AC component of the first signal component, Δsig2 represents the second signal component, and f d2 represents the second DC component of the second signal component, f s2 a second AC component representing a second signal component;

[0073] Step S2: Obtain the green light filtering spectrum matrix r of the green light filtering mechanism, and calculate the first signal component Δsig1 (f d1 ,f s1 ) and the second signal component Δsig2(f d2 ,f s2) performing cross spectrum calculation to determine the cross signal;

[0074]

[0075] Among them, the S n (sig) represents the cross signal, E represents the unit matrix with the same factorial as the green light filter spectrum matrix, i represents the ordinal number of the i-th signal component collected in the first signal component and the second signal component, n represents the first signal component and the second signal component collected in total, i∈(1,n);

[0076] Step S3: transmitting the cross signal to the green light filtering mechanism to obtain a dynamic adaptive model function;

[0077]

[0078] Wherein, the fer dynamic adaptive model function, T represents all signal cycles collected, t represents a complete signal cycle in the complete signal cycle, represents the dynamic adaptive function, θ represents the dynamic different signal period deviations based on the adaptive situation, and q represents the number of different signal period deviations collected;

[0079] Step S4: When the fer dynamic adaptive model function cannot be synthesized, the wearable device automatically reports an error;

[0080] Step S5: When the fer dynamic adaptive model function is synthesized, a dynamic adaptive model is established through the dynamic adaptive model function.

[0081] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0082] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0084] Figure 1 This is a flow chart of a three-wavelength blood oxygen saturation monitoring method based on a wearable device in an embodiment of the present invention;

[0085] Figure 2This is a flow chart of a three-wavelength blood oxygen saturation monitoring method based on a wearable device in an embodiment of the present invention;

[0086] Figure 3 This is a flow chart of a three-wavelength blood oxygen saturation monitoring method based on a wearable device in an embodiment of the present invention. DETAILED DESCRIPTION

[0087] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0088] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0089] Embodiment 1:

[0090] according to Figure 1 As shown, the present invention provides a three-wavelength blood oxygen saturation monitoring method based on a wearable device, characterized in that it includes:

[0091] Acquire the pulse wave signal of the object to be detected; wherein,

[0092] The pulse wave signal includes a green light pulse wave signal, a red light pulse wave signal and an infrared light pulse wave signal;

[0093] Based on the environment of the object to be detected, a state filtering range of the green light pulse wave signal is obtained, and a green light filtering mechanism is generated according to the state filtering range;

[0094] According to the red light pulse wave signal, searching for red light characteristic points, and determining a first signal component based on the red light characteristic points;

[0095] According to the infrared light pulse wave signal, searching for infrared light characteristic points, and determining a second signal component based on the infrared light characteristic points;

[0096] The first signal component and the second signal component are filtered by a green light filtering mechanism, and a dynamic adaptive model is established after the processing;

[0097] The blood oxygen saturation of the object to be detected is calculated based on the dynamic adaptive model.

[0098] The working principle of the above technical solution is:

[0099] The embodiment of the present invention provides a three-wavelength blood oxygen saturation monitoring method based on a wearable device, which obtains the user's green light pulse wave signal by using a watch when the user is in a stationary state, and the watch issues a collection instruction to obtain the user's red light pulse wave signal and infrared light pulse wave signal, and obtains the pulse wave signal of the object to be detected when it is in a stationary state; wherein the pulse wave signal includes a green light pulse wave signal, a red light pulse wave signal and an infrared light pulse wave signal; then, according to the green light, a suitable filtering method is determined to perform spectrum analysis on the green light to determine a filtering frequency range suitable for the current state and environment, and according to the filtering range of the green light and the red light pulse wave signal, a red light feature point is searched, and based on the red light feature point, a first signal component is determined; according to the infrared light pulse wave signal, an infrared light feature point is searched, and based on the infrared light feature point, a second signal component is determined; the red light and infrared light pulse wave signals are processed, and then the red light pulse wave signals are calculated. DC component and red light AC component, and find the red light feature point, calculate the red light DC component and red light AC component according to a preset formula, then calculate the infrared light DC component and infrared light AC component, find the infrared light feature point, calculate the infrared light DC component and infrared light AC component according to the formula, calculate the blood oxygen saturation, calculate the wavelength coefficient ratio of the calculated result, and use a linear equation to calculate the blood oxygen saturation. The green light signal of traditional wearable devices is better than the red light and infrared light signals, and can be used to establish a suitable filtering spectrum range and method. The filtering spectrum range is closely related to the current environment and state, so it has a certain degree of adaptability. The whole process obtains the state filtering range of the green light pulse wave signal, and generates a green light filtering mechanism according to the state filtering range; based on the green light filtering mechanism, the wavelength coefficient ratio is calculated through the first signal component and the second signal component, and the blood oxygen saturation of the object to be detected is determined according to the wavelength coefficient ratio.

[0100] The beneficial effects of the above technical solution are:

[0101] This technical solution obtains the entire filtering spectrum range of the current environment and state through the adaptability based on green light, and obtains the user's pulse signal based on the green light filtering mechanism under the irradiation of red light and infrared light based on the user's pulse, thereby determining the blood oxygen concentration and ensuring the user's health.

[0102] Embodiment 2:

[0103] The technical solution provides an embodiment, before obtaining the pulse wave signal of the object to be detected, it also includes:

[0104] Based on the preset sensing device in the wearable device, determine whether the object to be detected is in a stationary state:

[0105] When it is detected that the object to be detected is in a stationary state, based on the preset wearable device, a collection instruction is automatically issued to collect the red light pulse wave signal and infrared light pulse signal of the object to be detected;

[0106] When it is detected that the object to be detected is not in a static state, based on the preset wearable device, a collection instruction is automatically issued to collect the green light pulse wave signal of the object to be detected.

[0107] The working principle of the above technical solution is:

[0108] The present technical solution provides an embodiment, which includes determining whether the object to be detected is in a stationary state before acquiring the pulse wave signal when the object to be detected is in a stationary state, because the pulse of a person in different states and environments is different and will have different influences, so it is necessary to first obtain the green light filter value of the object to be detected, that is, the green light pulse signal of the user in a large environment and in a stationary state, and then when it is detected that the object to be detected is in a stationary state, based on a preset wearable device, an acquisition instruction is automatically issued to acquire the pulse wave signal of the object to be detected; when it is detected that the object to be detected is not in a stationary state, the wearable device automatically enters a power saving mode.

[0109] The beneficial effects of the above technical solution are:

[0110] When the green light filter value is monitored when the user is in a stationary state, this technical solution has an environmental spectrum collection operation for adaptively adapting to the surrounding environmental spectrum, thereby making the collected pulse signal data more accurate and more adapted to the current environmental state.

[0111] Embodiment 3:

[0112] The technical solution provides an embodiment, wherein the step of obtaining a pulse wave signal of a subject to be detected when the subject is in a stationary state comprises:

[0113] Based on the preset wearable device, light waves are emitted alternately to the object to be detected in a stationary state at regular intervals; wherein,

[0114] The light wave rays include green light rays, red light rays and infrared light rays;

[0115] Acquiring the light intensity of the red light and the infrared light, and determining the arterial pulsation optical path of the object to be detected according to the light intensity;

[0116] According to the arterial pulsation optical path, the pulse intensity of the object to be detected is recorded in real time;

[0117] A pulse wave signal of the object to be detected is determined according to the pulse intensity.

[0118] The working principle of the above technical solution is:

[0119] The technical solution obtains the detection object in a static state and compares it with the detection object in a non-static state, and after a large amount of data recognition, obtains the pulse wave signal when the detection object is in a static state, and based on the preset wearable device, regularly emits light waves to the object to be detected in a static state alternately; wherein the light waves include green light, red light and infrared light; the light intensity of the red light and the infrared light is obtained, and the arterial pulsation optical path of the object to be detected is determined according to the light intensity; according to the arterial pulsation optical path, the pulse intensity of the object to be detected is recorded in real time; according to the pulse intensity, the pulse wave signal of the object to be detected is determined, and the user's pulse is measured through three different lights, so as to obtain the pulse signal of the user in different environmental states, collect and analyze the spectrum of the surrounding environment, and determine the accurate pulse signal.

[0120] The beneficial effects of the above technical solution are:

[0121] This technical solution acquires the green light through the collection, and establishes an adaptive green light filtering mechanism according to the user status and the surrounding environment, so that the collection of the user's pulse signal is more suitable for different occasions, different states and different locations, and the collected data is more accurate.

[0122] Embodiment 4:

[0123] according to Figure 2 As shown, the technical solution provides an embodiment, which obtains the state filtering range of the green light pulse wave signal based on the environment of the object to be detected, and generates a green light filtering mechanism according to the state filtering range, including:

[0124] acquiring a green light pulse wave signal of the object to be detected at a regular interval, and determining a characteristic frequency of the green light pulse wave signal according to the green light pulse wave signal;

[0125] According to the characteristic frequency, the signal influencing parameter of the pulse wave signal is obtained; wherein,

[0126] The signal influencing parameters include the fluctuation coefficient, cut-off frequency and low pass resistance order of the pulse wave signal;

[0127] According to the signal impact parameters, the maximum attenuation data is calculated; wherein,

[0128] The loss data is loss data under the influence of the wearable device's own parameters;

[0129] Based on the environment of the object to be detected, the maximum attenuation data is transmitted to a preset Chebyshev filter, and a green light filtering mechanism is established.

[0130] The working principle of the above technical solution is:

[0131] The present technical solution provides an embodiment, which is an operation process for obtaining a pulse wave signal of an object to be detected when it is in a stationary state. The wearable device of the present technical solution is used to periodically emit light waves alternately to the object to be detected when it is in a stationary state; wherein the light waves include green light, red light and infrared light; the light intensity of the red light and the infrared light is obtained, and the light intensity is based on the tissue thickness of the object to be detected and the concentration of the fluid in the body. The incident intensity of the light is determined by the conditions of the object to be detected and the speed of light propagation, and the arterial pulsation optical path of the object to be detected is determined according to the light intensity; the pulse intensity of the object to be detected is recorded in real time according to the arterial pulsation optical path; the pulse wave signal of the object to be detected is determined according to the pulse intensity, and the pulse wave signal here is the most original pulse signal, which is an adaptive spectrum generated by collecting environmental information based on the user's environment and state, such as intense exercise in a gym or in a special environment of a sauna in a sauna room.

[0132] The beneficial effects of the above technical solution are:

[0133] This technical solution provides an adaptive green light filtering mechanism, which monitors and collects the state of the user's environment in real time, and adapts through autonomous learning to determine a pulse that is more suitable for different places.

[0134] Embodiment 5:

[0135] according to Figure 3 As shown, the technical solution provides an embodiment, based on the environment of the object to be detected, the maximum attenuation data is transmitted to a preset Chebyshev filter, and a green light filtering mechanism is established, including:

[0136] Based on the environment of the object to be detected, the green light filtering frequency is collected through a preset Chebyshev filter; wherein,

[0137] The filtering frequency is the noise frequency and abnormal signal of the object to be detected in the environment;

[0138] Calculating the pulse wave characteristic matrix of the object to be detected in a non-stationary state;

[0139] Acquire and transmit the maximum attenuation data to the Chebyshev filter, and filter the pulse wave characteristic matrix through the green light filtering frequency to determine the target filtering matrix;

[0140] A green light filtering mechanism is established according to the target filtering matrix.

[0141] The working principle of the above technical solution is:

[0142] The technical solution provides an embodiment, wherein the state filtering range of a green light pulse wave signal is obtained, and a green light filtering mechanism is generated according to the state filtering range, including: regularly obtaining a green light pulse wave signal of an object to be detected, and determining a characteristic frequency of the green light pulse wave signal according to the green light pulse wave signal; obtaining a signal influencing parameter of the pulse wave signal according to the characteristic frequency; wherein the signal influencing parameter includes a fluctuation coefficient, a cutoff frequency, and a low pass resistance order of the pulse wave signal; calculating maximum attenuation data of a Chebyshev filter passband according to the signal influencing parameter; transmitting the maximum attenuation data to a preset Chebyshev filter, and establishing a green light filtering mechanism.

[0143] Embodiment 6:

[0144] The technical solution provides an embodiment, wherein the calculation of the pulse wave characteristic matrix of the object to be detected in a non-stationary state includes:

[0145] When the object to be detected is in a non-stationary state, a green light pulse wave signal of the object to be detected is collected;

[0146] Determining green light pulse wave characteristic points according to the green light pulse wave signal;

[0147] Fitting and calculating the pulse wave characteristic points, and determining the characteristic coefficients of the green light pulse characteristic points;

[0148] The pulse wave characteristic points are matrixed according to the characteristic coefficients to determine a pulse wave characteristic matrix.

[0149] The working principle and beneficial effects of the above technical solution are:

[0150] The technical solution provides an embodiment, wherein the maximum attenuation data is transmitted to a preset Chebyshev filter and a green light filtering mechanism is established, including: collecting a green light pulse wave signal through the Chebyshev filter and determining a green light pulse wave feature point of the green light pulse wave signal; fitting the pulse wave feature point to determine a feature coefficient of the green light pulse feature point, and matrixing the pulse wave feature point according to the feature coefficient to determine a pulse wave feature matrix; transmitting the maximum attenuation data to a preset Chebyshev filter, filtering the pulse wave feature matrix to determine a target filtering matrix; and establishing a green light filtering mechanism according to the target filtering matrix. The entire process is adaptive to the environment in which the object to be detected is located, thereby improving the detection accuracy of the object to be detected.

[0151] Embodiment 7:

[0152] The technical solution provides an embodiment, wherein the green light filtering mechanism is established according to the target filtering matrix, including:

[0153] Step 1: Get the preset ideal filter matrix

[0154]

[0155] Wherein, Density represents the ideal filter matrix, n×n represents the ideal filter matrix is ​​a matrix with n rows and n columns, and density represents the pulse wave characteristic data under the ideal filtering condition in the ideal filter matrix;

[0156] Step 2: Get the target filter matrix

[0157]

[0158] Wherein, target represents the target filter matrix, n×n represents the target filter matrix is ​​a matrix with n rows and n columns, and target represents the pulse wave characteristic data under the target filtering condition in the target filter matrix;

[0159] Step 3: Determine a transform coefficient matrix according to the target filter matrix and the ideal filter matrix;

[0160]

[0161] Among them, W n×n represents the transformation coefficient matrix, E n×n represents the identity matrix;

[0162] Step 4: Establish a green light filter spectrum matrix through the transformation coefficient matrix;

[0163]

[0164] Where r represents the green light filter spectrum matrix;

[0165] Step 5: Determine a green light filtering mechanism according to the green light filtering spectrum matrix.

[0166] The working principle of the above technical solution is:

[0167] This technical solution provides an embodiment to obtain a preset ideal filter matrix Density n×n , the ideal filter matrix contains the pulse wave characteristic data density under the ideal filtering condition, obtain the target filter matrix Target, the target filter matrix contains the pulse wave characteristic data target under the target filtering condition; according to the target filter matrix and the ideal filter matrix, determine the transformation coefficient matrix W n×n , which is to filter the ideal parameters of the ideal filter matrix through the target filter matrix, and determine the green light filter spectrum matrix r and the transformation coefficient matrix W n×nThe coefficient correlation matrix between them is used to determine the green light filtering mechanism according to the green light filtering spectrum matrix, so as to establish the green light filtering mechanism according to the target filtering matrix.

[0168] The beneficial effects of the above technical solution are:

[0169] This technical solution determines the transformation space matrix of the target filter matrix by converting the coefficients of the acquired target filter matrix under ideal conditions, which greatly expands the flexibility and robustness of the pulse feature point conversion and improves the fault tolerance and adaptability of the pulse feature points.

[0170] Embodiment 8:

[0171] The technical solution provides an embodiment, wherein the step of searching for red light feature points according to the red light pulse wave signal and determining a first signal component based on the red light feature points includes:

[0172] Calculating the red light pulse wave intensity according to the red light pulse wave signal;

[0173] Sorting the intensity of the red light pulse wave and determining a first intensity peak;

[0174] Collecting and reconstructing red light feature points through the first intensity peak;

[0175] According to the red light feature points, the red light DC component and the red light AC component are calculated; wherein,

[0176] The red light AC component is a change signal of arterial pulse obtained based on photoplethysmography under the action of red light;

[0177] A first signal component is determined according to the red light DC component and the red light AC component.

[0178] The working principle and beneficial effects of the above technical solution are:

[0179] The technical solution provides an embodiment, according to the red light pulse wave signal, the red light feature point is searched, and based on the red light feature point, the first signal component is determined, the first component is composed of the red light AC component and the red light DC component, the feature point refers to the regular rhythm of the user's body pulse under the irradiation of red light, the peak and trough values ​​of the regular rhythm are collected to form the feature point, according to the red light pulse wave signal, the red light pulse wave intensity is calculated, which is similar to the infrared light pulse wave intensity, and the red light pulse wave intensity is intensity-sorted, and the first intensity peak value is determined; through the first intensity peak value, the red light feature point is collected and reconstructed, because the rhythm of the red light feature point is complex The infrared light or red light is complex and changeable, and there are many noises or other interference factors, so it is necessary to shunt the infrared light or the red light. Direct current refers to the unchanged logistics signal, and alternating current refers to the signal with constantly changing direction. According to the feature points outside the red light, the DC component of the red light and the AC component of the red light are calculated; wherein, the AC component of the red light is the change signal of the arterial pulse obtained based on the photoelectric volumetric recording method under the action of the infrared light; according to the DC component of the red light and the AC component of the red light, the first signal component is determined. The entire embodiment analyzes the pulse wave under the irradiation of infrared light and red light, thereby slowly stripping off the rough signal collected and obtaining the most accurate feature points, and has excellent robustness and anti-interference capabilities.

[0180] Embodiment 9:

[0181] The technical solution provides an embodiment, wherein searching for infrared light feature points according to the infrared light pulse wave signal and determining a second signal component based on the infrared light feature points comprises:

[0182] Calculating the infrared light pulse wave intensity according to the infrared light pulse wave signal;

[0183] Sorting the intensity of the infrared light pulse wave and determining a second intensity peak;

[0184] Collecting and reconstructing infrared light feature points through the second intensity peak;

[0185] According to the red light external feature points, the infrared light DC component and the infrared light AC component are calculated; wherein,

[0186] The infrared light AC component is a change signal of the arterial pulse obtained based on photoplethysmography under the action of the infrared light;

[0187] A second signal component is determined according to the infrared light DC component and the infrared light AC component.

[0188] The working principle and beneficial effects of the above technical solution are:

[0189] The technical solution provides an embodiment, according to the infrared light pulse wave signal, the infrared light feature point is searched, and based on the infrared light feature point, the second signal component is determined, the second signal component is composed of the infrared light AC component and the infrared light DC component, the feature point refers to the regular rhythm of the user's body pulse under the irradiation of infrared light, the peak and trough values ​​of the regular rhythm are collected to form the feature point, according to the infrared light pulse wave signal, the infrared light pulse wave intensity is calculated, which is similar to the red light pulse wave intensity, and the infrared light pulse wave intensity is intensity-sorted, and the second intensity peak is determined; through the second intensity peak, the infrared light feature point is collected and reconstructed, because the infrared light feature point The rhythm is complex and changeable, and there are many noises or other interference factors, so it is necessary to shunt the infrared light or red light. Direct current refers to the unchanged logistics signal, and alternating current refers to the signal with constantly changing direction. According to the feature points outside the red light, the direct current component of the infrared light and the alternating current component of the infrared light are calculated; wherein, the alternating current component of the infrared light is a change signal of the arterial pulse obtained based on photoplethysmography under the action of the infrared light; according to the direct current component of the infrared light and the alternating current component of the infrared light, the second signal component is determined. The entire embodiment analyzes the pulse wave under the irradiation of infrared light and red light, thereby slowly stripping off the rough signal collected and obtaining the most accurate feature points, and has excellent robustness and anti-interference capabilities.

[0190] Embodiment 10:

[0191] The technical solution provides an embodiment, wherein the first signal component and the second signal component are filtered through a green light filtering mechanism, and a dynamic adaptive model is established after the processing, and the first signal component and the second signal component are filtered through a green light filtering mechanism, and a dynamic adaptive model is established after the processing, including:

[0192] Step S1: Collect the first signal component Δsig1 (f d1 ,f s1 ) and the second signal component Δsig2(f d2 ,f s2 );

[0193] Where Δsig1 represents the first signal component, f d1 represents the first DC component of the first signal component, f s1 represents the first AC component of the first signal component, Δsig2 represents the second signal component, and f d2 represents the second DC component of the second signal component, f s2 a second AC component representing a second signal component;

[0194] Step S2: Obtain the green light filtering spectrum matrix r of the green light filtering mechanism, and calculate the first signal component Δsig1 (fd1 ,f s1 ) and the second signal component Δsig2(f d2 ,f s2 ) performing cross spectrum calculation to determine the cross signal;

[0195]

[0196] Among them, the S n (sig) represents the cross signal, E represents the unit matrix with the same factorial as the green light filter spectrum matrix, i represents the ordinal number of the i-th signal component collected in the first signal component and the second signal component, n represents the first signal component and the second signal component collected in total, i∈(1,n);

[0197] Step S3: transmitting the cross signal to the green light filtering mechanism to obtain a dynamic adaptive model function;

[0198]

[0199] Wherein, the fer dynamic adaptive model function, T represents all signal cycles collected, t represents a complete signal cycle in the complete signal cycle, represents the dynamic adaptive function, θ represents the dynamic different signal period deviations based on the adaptive situation, and q represents the number of different signal period deviations collected;

[0200] Step S4: When the fer dynamic adaptive model function cannot be synthesized, the wearable device automatically reports an error;

[0201] Step S5: When the fer dynamic adaptive model function is synthesized, a dynamic adaptive model is established through the dynamic adaptive model function.

[0202] The working principle of the above technical solution is:

[0203] The technical solution collects and stores the first signal component and the second signal component, establishes a dynamic adaptive model based on the green light filtering mechanism, and determines the blood oxygen saturation of the object to be detected based on the dynamic adaptive model, including: collecting the first signal component Δsig1 (f d1 ,f s1 ) and the second signal component Δsig2(f d2 ,f s2 ); obtain the green light filtering spectrum matrix r of the green light filtering mechanism, and for the first signal component Δsig1(f d1 ,f s1 ) and the second signal component Δsig2(f d2 ,f s2) performs cross spectrum calculation to determine the cross signal; S n (sig) The cross signal is transmitted to the green light filtering mechanism to obtain a dynamic adaptive model function; when the fer dynamic adaptive model function cannot be synthesized, the wearable device automatically reports an error; when the fer dynamic adaptive model function is synthesized, a dynamic adaptive model is established through the dynamic adaptive model function; based on the dynamic adaptive model, the physical sign data of the object to be detected is obtained and processed to determine the blood oxygen saturation of the object to be detected.

[0204] The beneficial effects of the above technical solution are:

[0205] This technical solution establishes an adaptive model, continuously processes multiple shunt signals, removes impurities, and establishes an adaptive learning model to meet a variety of situations. It can flexibly meet the needs of the objects to be detected in different scenes and states, monitor the physical condition of the objects to be detected in real time, and has excellent robustness and anti-interference capabilities.

[0206] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0207] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0208] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0209] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0210] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A three-wavelength blood oxygen saturation monitoring method based on wearable devices, characterized in that: include: Acquire the pulse wave signal of the object to be detected; wherein, The pulse wave signal includes a green light pulse wave signal, a red light pulse wave signal and an infrared light pulse wave signal; Based on the environment of the object to be detected, a state filtering range of the green light pulse wave signal is obtained, and a green light filtering mechanism is generated according to the state filtering range; According to the red light pulse wave signal, searching for red light characteristic points, and determining a first signal component based on the red light characteristic points; According to the infrared light pulse wave signal, searching for infrared light characteristic points, and determining a second signal component based on the infrared light characteristic points; The first signal component and the second signal component are filtered by a green light filtering mechanism, and a dynamic adaptive model is established after the processing; Calculating the blood oxygen saturation of the object to be detected based on the dynamic adaptive model; The state filtering range of the green light pulse wave signal is obtained based on the environment of the object to be detected, and a green light filtering mechanism is generated according to the state filtering range, including: acquiring a green light pulse wave signal of the object to be detected at a regular interval, and determining a characteristic frequency of the green light pulse wave signal according to the green light pulse wave signal; According to the characteristic frequency, the signal influencing parameter of the pulse wave signal is obtained; wherein, The signal influencing parameters include the fluctuation coefficient, cut-off frequency and low pass resistance order of the pulse wave signal; According to the signal impact parameters, the maximum attenuation data is calculated; wherein, The maximum attenuation data is based on the loss data under the influence of the wearable device's own parameters; Based on the environment of the object to be detected, the maximum attenuation data is transmitted to a preset Chebyshev filter, and a green light filtering mechanism is established.

2. A three-wavelength blood oxygen saturation monitoring method based on a wearable device as claimed in claim 1, characterized in that: Before acquiring the pulse wave signal of the object to be detected, the method further includes: Based on the preset sensing device in the wearable device, determine whether the object to be detected is in a stationary state: When it is detected that the object to be detected is in a stationary state, based on the preset wearable device, a collection instruction is automatically issued to collect the red light pulse wave signal and infrared light pulse signal of the object to be detected; When it is detected that the object to be detected is not in a static state, based on the preset wearable device, a collection instruction is automatically issued to collect the green light pulse wave signal of the object to be detected.

3. A three-wavelength blood oxygen saturation monitoring method based on a wearable device as claimed in claim 1, characterized in that: The step of obtaining a pulse wave signal when the object to be detected is in a stationary state comprises: Based on the preset wearable device, light waves are emitted alternately to the object to be detected in a stationary state at regular intervals; wherein, The light wave rays include green light rays, red light rays and infrared light rays; Acquiring the light intensity of the red light and the infrared light, and determining the arterial pulsation optical path of the object to be detected according to the light intensity; According to the arterial pulsation optical path, the pulse intensity of the object to be detected is recorded in real time; A pulse wave signal of the object to be detected is determined according to the pulse intensity.

4. A three-wavelength blood oxygen saturation monitoring method based on a wearable device as claimed in claim 1, characterized in that: Based on the environment of the object to be detected, the maximum attenuation data is transmitted to a preset Chebyshev filter, and a green light filtering mechanism is established, including: Based on the environment of the object to be detected, the green light filtering frequency is collected through a preset Chebyshev filter; wherein, The filtering frequency is the noise frequency and abnormal signal of the object to be detected in the environment; Calculating the pulse wave characteristic matrix of the object to be detected in a non-stationary state; Acquire and transmit the maximum attenuation data to the Chebyshev filter, and filter the pulse wave characteristic matrix through the green light filtering frequency to determine the target filtering matrix; A green light filtering mechanism is established according to the target filtering matrix.

5. A three-wavelength blood oxygen saturation monitoring method based on a wearable device as claimed in claim 4, characterized in that: The step of calculating the pulse wave characteristic matrix of the object to be detected in a non-stationary state includes: When the object to be detected is in a non-stationary state, a green light pulse wave signal of the object to be detected is collected; Determining green light pulse wave characteristic points according to the green light pulse wave signal; Fitting and calculating the pulse wave characteristic points, and determining the characteristic coefficients of the green light pulse characteristic points; The pulse wave characteristic points are matrixed according to the characteristic coefficients to determine a pulse wave characteristic matrix.

6. A three-wavelength blood oxygen saturation monitoring method based on a wearable device as claimed in claim 4, characterized in that: The step of establishing a green light filtering mechanism according to the target filtering matrix includes: Step 1: Get the preset ideal filter matrix Wherein, Density represents the ideal filter matrix, n×n represents the ideal filter matrix is ​​a matrix with n rows and n columns, and density represents the pulse wave characteristic data under the ideal filtering condition in the ideal filter matrix; Step 2: Get the target filter matrix Among them, target represents the target filter matrix, n×n represents the target filter matrix is ​​a matrix with n rows and n columns, and target represents the pulse wave feature data under the target filtering condition in the target filter matrix; Step 3: Determine a transform coefficient matrix according to the target filter matrix and the ideal filter matrix; Among them, W n×n represents the transformation coefficient matrix, E n×n represents the identity matrix; Step 4: Establish a green light filter spectrum matrix through the transformation coefficient matrix; Where r represents the green light filter spectrum matrix; Step 5: Determine a green light filtering mechanism according to the green light filtering spectrum matrix.

7. A three-wavelength blood oxygen saturation monitoring method based on a wearable device as claimed in claim 1, characterized in that: The step of searching for red light feature points according to the red light pulse wave signal and determining a first signal component based on the red light feature points includes: Calculating the red light pulse wave intensity according to the red light pulse wave signal; Sorting the intensity of the red light pulse wave and determining a first intensity peak; Collecting and reconstructing red light feature points through the first intensity peak; According to the red light feature points, the red light DC component and the red light AC component are calculated; wherein, The red light AC component is a change signal of arterial pulse obtained based on photoplethysmography under the action of red light; A first signal component is determined according to the red light DC component and the red light AC component.

8. A three-wavelength blood oxygen saturation monitoring method based on a wearable device as claimed in claim 1, characterized in that: The step of searching for infrared light characteristic points according to the infrared light pulse wave signal and determining a second signal component based on the infrared light characteristic points includes: Calculating the infrared light pulse wave intensity according to the infrared light pulse wave signal; Sorting the intensity of the infrared light pulse wave and determining a second intensity peak; Collecting and reconstructing infrared light feature points through the second intensity peak; According to the red light external feature points, the infrared light DC component and the infrared light AC component are calculated; wherein, The infrared light AC component is a change signal of the arterial pulse obtained based on photoplethysmography under the action of the infrared light; A second signal component is determined according to the infrared light DC component and the infrared light AC component.

9. A three-wavelength blood oxygen saturation monitoring method based on a wearable device as claimed in claim 1, characterized in that: The first signal component and the second signal component are filtered through a green light filtering mechanism, and a dynamic adaptive model is established after the processing. The first signal component and the second signal component are filtered through a green light filtering mechanism, and a dynamic adaptive model is established after the processing. include: Step S1: Collect the first signal component Δsig1 (f d1 , f s1 ) and the second signal component Δsig2(f d2 , f s2 ); Where Δsig1 represents the first signal component, f d1 represents the first DC component of the first signal component, f s1 represents the first AC component of the first signal component, Δsig2 represents the second signal component, and f d2 represents the second DC component of the second signal component, f s2 a second AC component representing a second signal component; Step S2: Obtain the green light filtering spectrum matrix r of the green light filtering mechanism, and calculate the first signal component Δsig1 (f d1 , f s1 ) and the second signal component Δsig2(f d2 , f s2 ) performing cross spectrum calculation to determine the cross signal; Among them, the S n (sig) represents the cross signal, E represents the unit matrix with the same factorial as the green light filter spectrum matrix, i represents the ordinal number of the i-th signal component collected in the first signal component and the second signal component, n represents the first signal component and the second signal component collected in total, i∈(1,n); Step S3: transmitting the cross signal to the green light filtering mechanism to obtain a dynamic adaptive model function; Wherein, the fer dynamic adaptive model function, T represents all signal cycles collected, t represents a complete signal cycle in the complete signal cycle, represents the dynamic adaptive function, θ represents the dynamic different signal period deviations based on the adaptive situation, and q represents the number of different signal period deviations collected; Step S4: When the fer dynamic adaptive model function cannot be synthesized, the wearable device automatically reports an error; Step S5: When the fer dynamic adaptive model function is synthesized, a dynamic adaptive model is established through the dynamic adaptive model function.

Citation Information

Patent Citations

  • Method for eliminating interference to blood oxygen saturation monitoring

    CN102499694A

  • Driving fatigue reminding bracelet with intelligent physiological fatigue detection function

    CN109044296A

  • Health monitoring systems and methods

    CN110087543A

  • Blood oxygen detection method and device

    CN111685726A

  • Multi-wavelength pulse oximetry

    US10568525B1