CN119257595A

CN119257595A - A non-invasive blood oxygen detection method

Info

Publication number: CN119257595A
Application number: CN202411574885.6A
Authority: CN
Inventors: 刘远全
Original assignee: Chongqing College of Finance and Economics
Current assignee: Chongqing College of Finance and Economics
Priority date: 2024-11-06
Filing date: 2024-11-06
Publication date: 2025-01-07

Abstract

The invention belongs to the technical field of blood oxygen detection, and particularly relates to a noninvasive blood oxygen detection method, which comprises a blood oxygen machine and a main machine shell, wherein the main machine shell is arranged at the upper part of the blood oxygen machine, a blood oxygen sensor is arranged below the blood oxygen machine and is electrically connected with the main machine shell, the main machine shell comprises a singlechip, an analog front end driving module, a display module, a circuit module, a signal receiving module, a digital signal processing module and a storage module, meanwhile, a lithium battery is arranged in the main machine shell and is used for providing power for the blood oxygen machine, the blood oxygen sensor is directly connected with the signal receiving module to form a blood oxygen signal receiving module, the blood oxygen signal receiving module is connected with a digital signal processing module to convert a voltage signal into a digital signal, the problems that a detection signal pulse wave output signal is subjected to serious noise interference such as environmental noise and a hardware circuit are solved, and the problems of signal loss and signal superposition caused by a traditional filtering method are solved.

Description

Noninvasive blood oxygen detection method

Technical Field

The invention relates to the technical field of blood oxygen detection, in particular to a noninvasive blood oxygen detection method.

Background

Blood oxygen saturation (SpO ₂), the most important indicator in blood oxygen testing, represents the percentage of the volume of oxygenated hemoglobin (HbO ₂) in the blood that is bound by oxygen over the volume of total hemoglobin (Hb) that can be bound. The blood oxygen saturation of normal human arterial blood is generally between 95% and 98%, and venous blood is about 75%;

Blood oxygen saturation is an important physiological parameter reflecting respiratory and circulatory functions. If the blood oxygen saturation is lower than the normal range, the organism may be prompted to have an anoxic condition, such as lung diseases (pneumonia, chronic obstructive pulmonary disease and the like), cardiovascular diseases (heart failure and the like) or altitude reactions and the like;

however, the existing blood oxygen detection has some problems, such as inaccurate detection, more interference, such as environmental interference, hardware circuit noise interference, and the like, and the traditional filtering method is easy to cause the problems of signal loss and signal superposition, causes signal receiving distortion, and can adversely affect the detection result.

Disclosure of Invention

The invention aims to provide a noninvasive blood oxygen detection method for solving the problems in the background technology.

In order to achieve the aim, the invention provides the following technical scheme that the noninvasive blood oxygen detection and method comprises a blood oxygen machine and a main machine shell;

The main machine shell is arranged at the upper part of the blood oxygen machine, a blood oxygen sensor is arranged below the blood oxygen machine, the blood oxygen sensor is electrically connected with the main machine shell, the main machine shell comprises a singlechip, an analog front end driving module, a display module, a circuit module, a signal receiving module, a digital signal processing module and a storage module, and meanwhile, a lithium battery is arranged in the main machine shell to provide power for the blood oxygen machine;

The blood oxygen sensor is directly connected with the signal receiving module to form a blood oxygen signal receiving module, and the blood oxygen signal receiving module is connected with the digital signal processing module to convert the voltage signal into a digital signal;

The signals collected by the blood oxygen machine comprise electrocardiosignals, photoelectric volume pulse wave signals and skin temperature signals, and the detection method of the blood oxygen machine comprises the following steps:

preprocessing the acquired physiological signals to remove noise and interference;

Extracting characteristic parameters of electrocardiosignals and photoelectric volume pulse wave signals, including heart rate, pulse wave amplitude and pulse wave conduction time;

according to the extracted characteristic parameters, calculating the blood oxygen saturation by adopting a PPG measurement method;

And the blood oxygen detection is optimized, and the calculated blood oxygen saturation is corrected by combining the skin temperature signal, so that the detection accuracy is improved.

Preferably, the analog front end driving module uses five connection ports including a power interface VIN (3.3V/5V), a ground port GND, a clock interface SCL, a data interface SDA, and an interrupt interface INT, where the clock interface SCL is connected to the single chip microcomputer port PB6, the data interface SDA is connected to the port PB7, and the interrupt interface INT is connected to the port PC8.

Preferably, the analog front end driving module comprises an integrated blood oxygen signal collector and an I2C communication module, the integrated blood oxygen signal collector adopts a miniature 5.6mm x 3.3mm x 1.55mm 14-needle light collecting module, the analog front end driving module comprises a subsystem, and the subsystem comprises ambient light elimination and a proprietary discrete time filter.

Preferably, the blood oxygen detection method is optimized, and the steps are as follows:

S1, collecting at least three different wavelengths of light, irradiating a finger (at least three groups of different fingers) and an earlobe at a measured position, and synchronously collecting transmitted or reflected light signals, wherein the absorption characteristics of the different wavelengths of light on oxygenated hemoglobin and reduced hemoglobin are different, and the multi-wavelength light signal collection can provide richer information;

And S2, carrying out self-adaptive filtering processing on the collected optical signals, wherein a self-adaptive filtering algorithm automatically adjusts parameters of a filter according to the real-time characteristics of the signals.

S3, measuring some basic physiological characteristics of the tested person, such as skin color, tissue thickness and blood components, before testing, establishing a personalized light absorption correction model according to the individual physiological characteristics, and optimizing test data according to the light absorption correction model.

Preferably, according to said step S2, the high frequency noise component associated with the motion is identified by analyzing the frequency and amplitude variation characteristics of the optical signal, and then filtered out.

Preferably, according to the step S3, for the darker skin, when calculating the blood oxygen saturation, the light absorption coefficient is adjusted accordingly to compensate for the influence of melanin on light absorption.

Compared with the prior art, the invention has the beneficial effects that:

The invention solves the problem that the pulse wave output signal of the detection signal is interfered by serious noise such as environmental noise, hardware circuits and the like, adopts a hardware circuit design mode of combining an HTF52352 as a core processor chip and a blood oxygen signal acquisition analog front end MAX30102, reduces the complexity of circuit design, adopts differential transmission design for blood oxygen input signal input and increases shielding protection when designing a hardware circuit compared with the traditional circuit, can effectively avoid electromagnetic interference and improves the stability of the signal;

The method solves the problems of signal loss and signal superposition caused by the traditional filtering method, researches and improves a signal preprocessing method and a signal characteristic point detection algorithm, and according to the acquired original pulse wave signals, preprocessing waveforms in a mode of combining wavelet transformation and FFT aiming at the problems of signal data loss and distortion possibly occurring in the filtering process, and reconstructing the signals to remove the noise such as power frequency interference, baseline drift and the like in the original signal waveforms to a certain extent.

Drawings

FIG. 1 is a general block diagram of a blood oxygen signal acquisition module;

FIG. 2 is a schematic diagram of a monolithic circuit according to the present invention;

FIG. 3 is a schematic view of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

In the description of the present invention, it should be understood that the terms "upper," "lower," "front," "rear," "left," "right," "top," "bottom," "inner," "outer," and the like indicate or are based on the orientation or positional relationship shown in the drawings, merely to facilitate description of the present invention and to simplify the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention.

Examples:

Referring to fig. 1-3, the present invention provides a non-invasive blood oxygen detection method and technical scheme thereof, which comprises a blood oxygen machine 1 and a main machine shell 2;

the main machine shell 2 is arranged at the upper part of the blood oxygen machine 1, a blood oxygen sensor is arranged below the blood oxygen machine 1, the blood oxygen sensor is electrically connected with the main machine shell 2, the main machine shell 2 comprises a singlechip, an analog front end driving module, a display module, a circuit module, a signal receiving module, a digital signal processing module and a storage module, and meanwhile, a lithium battery is arranged in the main machine shell 2 to provide power for the blood oxygen machine 1;

All the components in the main shell 2 are connected through a data transmission line, and a shielding layer is arranged outside the data transmission line;

The analog front end driving module comprises an integrated blood oxygen signal collector and an I2C communication module, wherein the integrated blood oxygen signal collector adopts a miniature 5.6mm x 3.3mm x 1.55mm 14 needle light collection module; the analog front end driving module comprises a subsystem, wherein the subsystem comprises an ambient light elimination and a proprietary discrete time filter;

The analog front end driving module uses five wiring ports including a power interface VIN (3.3V/5V), a ground port GND, a clock interface SCL, a data interface SDA and an interrupt interface INT, wherein the clock interface SCL is connected with a singlechip port PB6, the data interface SDA is connected with a port PB7, and the interrupt interface INT is connected with a port PC8;

the acquisition signals of the blood oxygen machine 1 comprise electrocardiosignals, photoelectric volume pulse wave signals and skin temperature signals;

the blood oxygen detection method comprises the following steps:

correcting the calculated blood oxygen saturation by combining the skin temperature signal, so as to improve the detection accuracy;

The singlechip in the scheme adopts HT32F52352 singlechip, has the advantages of good electromagnetic interference resistance, excellent shading performance, high accuracy and the like, and comprises an Arm Cortex-M0+ processor, a bus structure and a memory system; The processor is a low-gate-number high-performance processor, is suitable for single-chip microcomputer and deep embedded application requiring area optimization and low-power-consumption processor, is based on an ARMv6-M structure and simultaneously supports Thumb instruction sets, single-cycle I/O ports, hardware multipliers and low-delay interrupt response time, is suitable for market products requiring high-performance and low-power-consumption single-chip microcomputer due to integration and advanced characteristics, and is mainly characterized in that I2C data read-write operation between a main chip and an analog front end MAX30102 is mainly completed through a microcontroller MCU (Micro control unit) in the processes of acquiring pulse wave data and processing, and a single-chip microcomputer circuit is shown in a figure 2;

Meanwhile, an SpO2 subsystem in the MAX30102 singlechip comprises ambient light elimination (ALC), a continuous-time sigma-delta ADC and a special discrete-time filter, wherein the ALC is provided with an internal track/hold circuit which can eliminate the ambient light and increase the effective dynamic range, the SpO2 ADC is provided with a programmable full-scale range of 2 mu A to 16 mu A, the ALC can eliminate the ambient current of up to 200 mu A, the internal ADC is a continuous-time oversampling sigma-delta converter with 18-bit resolution, the ADC sampling rate is 10.24MHz, the ADC output data rate can be programmed from 50sps (sampling per second) to 3200sps, and the overall structure of the blood oxygen signal acquisition module is shown in a figure 1;

Principle of PPG measurement:

A typical PPG system consists mainly of a light emitting diode LED, a photodiode PD, an analog front end AFE. First the LED emits incident light to the skin, which transmitted light is received by the PD and converted into a current signal, which is then converted into a digital quantity by the AFE. According to Beer-Lambert law, the absorptivity of light is different for different substances in blood vessels, and along with the beating of heart, the volume and internal solubles of blood vessels are also changed periodically, so that quantitative research can be carried out by using an optical method according to Beer-Lambert law, and further life indexes such as heart rate, blood oxygen, respiration and the like are extracted through PPG waveforms.

Beer-Lambert law describes the attenuation of light as a function of the characteristics of a substance through which the light passes:

I=I₀e^-ε(λ)Cd

Wherein A is attenuation, lambdao is incident light intensity, I is transmitted (received) light intensity, E (lambdax) is molar extinction coefficient of the substance, C is substance concentration, and d is light path length;

Blood oxygen saturation measurement principle:

Using Beer-Lambert law, blood oxygen is calculated using an optical method, where we define a quotient of the perfusion index (pi=ac/DC) of the PPG signal with R value of red and the PPG signal with infrared, then blood oxygen can be expressed as a quadratic function of blood oxygen, where the coefficients a, b, c of the quadratic function are related to the optical structure of the system, and we can be calibrated by fitting the calculated R value and the result of blood oxygen output by the standard device as reference to each other.

The formula is defined:

the calculation formula:

Wherein,

(The calibration coefficients a, b, c are related to the structure of the coefficient optics, which can be derived by fitting); the principle of discrete fourier transformation:

lemma 1 (elimination of lemma):

Lemma 2 (halving lemma):

At the position of Value y at ₀,y₁,...,y_n-1

Fast fourier transform (Fast Fourier Transform)

First, for this n-1 polynomial, its coefficient vector is a

We divide it into two vectors, even term and odd term

Even number:

odd number:

Their corresponding two polynomials are A ^[0] (x) and A ^[1] (x), and x is now substituted into

A(x)=a₀+a₁x+a₂x²+…+a_n-1x^n-1

A^[0](x)＝a₀+a₂x+a₄x²+…+a_n-2x^n/2-1

A^[1](x)＝a₁+a₃x+a₅x²+...+a_n-1x^n/2-1

The independent variable x of the two expressions is replaced by x ²

A^[0](x²)＝a₀+a₂x²+a₄x⁴+...+a_n-2x^n-2

A^[1](x²)＝a₁+a₃x²+a₅x⁴+...+a_n-1x^n-2

We multiply a ^[1] (x 2) by x to get

xA^[1](x3)＝a₁x+a₃x³+a₅x⁵+…+a_n-1x^n-1

Finally we can find that

A(x)=A^[0](x²)+xA^[1](x²)

At this time we convert the original problem, the value of A (x) on each unit root, into two polynomials of degree n/2, and the values of A ^[0] (x) and A ^[1] (x) on each unit root square are recombined. We substitute two specific unit roots, which are available according to lemma 1 and lemma 2,

According to the following:

The method is simplified to obtain the product,

Both A ^[0] (x) and A ^[1] (x) are exactly polynomials with the degree of n/2, the problem is converted into values of polynomials A ^[0] and A ^[1] with the degree of n/2 on each n/2-degree root, and the original problem is reduced by half.

By recursively solving the two sub-problems, the results can be quickly combined by halving the quotients. According to this idea we can run a recursive formula of the time T (n).

The total time is equal to 2 times of the combination time of the sub problem time plus O (n), and the time complexity is O (nlogn) according to the main theorem and the recursion tree;

the following is a pseudo code implementation of the FFT algorithm:

a_k＝ak_[0]+t

butterfly transformation:

if lim==1 return

a^[0]＝(a₀,a₂,…,a_n-2)

a^[1]＝(a₁,a₃,…,a_n-1)

FFT(a^[0],lim>>1)

FFT(a^[1],lim>>1)

ω=1

for k=0..n/2-1

a_k＝ak_[0]+t

ω=ωω_n

Wavelet transformation principle:

Wavelets, i.e. waves that exist in a small area. Let ψ (t) be a square integrable function, i.e. ψ (t) ε R if its Fourier transform ψ (ω) satisfies the condition

Ψ (t) is referred to as a basic wavelet, also known as the wavelet mother function. The wavelet mother function psi (t) is stretched and translated, the stretching factor (scale factor) is assumed to be a, the translation factor is tau, the function after stretching and translation is recorded as psi _a,τ (t), the real parameter pair psi _a,τ (t) is taken, the real parameter pair (a, tau) is taken, and the wavelet function is obtained:

The wavelet basis functions of parameters a and τ are called ψ _a,τ (t), because the values of both parameters a and τ change continuously. Therefore, it is also called continuous wavelet base transform. The arbitrary function f (t) of L ² (R) is unfolded under the wavelet base, and the continuous wavelet transformation of the function f (t) is called as:

WT _f (a, τ) is called a wavelet transform coefficient, whose inverse transform exists when the allowable conditions of the wavelet used are satisfied. The method comprises the following steps:

Discrete wavelet transform. Particularly, when a computer is used to perform a specific operation on signal data, the discrete processing of the continuous wavelet is a step that must be performed, and a specific discrete manner is composed of two steps, firstly, the discrete of the scale factor a, and the discrete method usually adopts binary discrete, namely a _m＝2^m. Wherein m is an integer. Next, the translation factor τ is discretized, typically by selecting discrete values that are equally spaced. The discrete mode follows the integral multiple of a _m, takes the minimum interval of the translation factor tau as T _s, and satisfies tau _k＝ka_mT_s

The wavelet function is expressed as:

the wavelet transform is expressed as:

From the above, it is clear that the continuous wavelet transform is unified in form with the discrete wavelet transform. In a specific computer operation process, the signals take discrete signals as main expression forms, so that the discrete wavelet transformation is selected as a main method when the signals are processed.

The blood oxygen detection is optimized so as to improve the accuracy of the detection, and the method comprises the following steps:

S1, irradiating fingers and earlobes of a measured part by adopting at least three different wavelengths of light, and synchronously collecting transmitted or reflected light signals, wherein the absorption characteristics of the different wavelengths of light on oxyhemoglobin and reduced hemoglobin are different, and the collection of the multi-wavelength light signals can provide richer information;

s2, carrying out self-adaptive filtering processing on the collected optical signals, wherein a self-adaptive filtering algorithm automatically adjusts parameters of a filter according to real-time characteristics of the signals, specifically, identifying high-frequency noise components related to movement by analyzing frequency and amplitude change characteristics of the optical signals, and filtering the high-frequency noise components;

S3, measuring some basic physiological characteristics of a tested person, such as skin color, tissue thickness and blood components, before detection, establishing a personalized light absorption correction model according to the individual physiological characteristics, and optimizing detection data according to the light absorption correction model;

For individuals with darker complexion, when calculating the blood oxygen saturation, the light absorption coefficient is correspondingly adjusted to compensate the influence of melanin on light absorption.

While the basic principles and main features of the present invention and advantages of the present invention have been shown and described above, it will be apparent to those skilled in the art that the present invention is not limited to the details of the above-described exemplary embodiments and can be embodied in other specific forms without departing from the spirit or essential features of the present invention, and therefore, the embodiments should be considered exemplary and non-limiting in all respects, the scope of the present invention is defined by the appended claims rather than the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.

Claims

1. A noninvasive blood oxygen detection method comprises a blood oxygen machine (1) and a main machine shell (2), and is characterized in that,

The host shell (2) is arranged at the upper part of the blood oxygen machine (1), a blood oxygen sensor is arranged below the blood oxygen machine (1), the blood oxygen sensor is electrically connected with the host shell (2), the host shell (2) comprises a singlechip, an analog front end driving module, a display module, a circuit module, a signal receiving module, a digital signal processing module and a storage module, and meanwhile, a lithium battery is arranged in the host shell (2) to provide power for the blood oxygen machine (1);

The blood oxygen sensor is directly connected with the signal receiving module to form a blood oxygen signal receiving module, and the blood oxygen signal receiving module is connected with the digital signal processing module to convert a voltage signal into a digital signal;

the signals collected by the blood oxygen machine (1) comprise electrocardiosignals, photoelectric volume pulse wave signals and skin temperature signals, and the detection method of the blood oxygen machine (1) comprises the following steps:

2. The method of claim 1, wherein the analog front end driver module uses five connection ports including a power interface VIN (3.3V/5V), a ground port GND, a clock interface SCL, a data interface SDA, and an interrupt interface INT, wherein the clock interface SCL is connected to a single chip microcomputer port PB6, the data interface SDA is connected to a port PB7, and the interrupt interface INT is connected to a port PC8.

3. The noninvasive blood oxygen detection and method according to claim 1, wherein the analog front end driving module comprises an integrated blood oxygen signal collector and an I2C communication module, the integrated blood oxygen signal collector adopts a miniature 5.6mm x 3.3mm x 1.55mm 14-needle light collection module, and the analog front end driving module comprises a subsystem, wherein the subsystem comprises ambient light elimination and a proprietary discrete-time filter.

4. The method for noninvasive blood oxygen detection according to claim 1, wherein the blood oxygen detection method is optimized, and the method comprises the following steps:

5. The method according to claim 4, wherein the step S2 is performed by analyzing the frequency and amplitude variation characteristics of the optical signal to identify and filter out high frequency noise components associated with the movement.

6. The method according to claim 4, wherein the step S3 is performed to compensate for the influence of melanin on the absorption of light by the dark-colored individual by adjusting the light absorption coefficient when calculating the blood oxygen saturation.