Blood oxygen saturation detection method
By combining a triaxial accelerometer with singular value decomposition processing of red and infrared PPG signals, the problems of insufficient accuracy and noise interference in blood oxygen saturation detection are solved, achieving more accurate blood oxygen saturation detection.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-09
- Publication Date
- 2026-04-10
AI Technical Summary
Existing methods for detecting blood oxygen saturation suffer from insufficient accuracy and significant noise interference, making it difficult to accurately reflect the binding of oxygen and hemoglobin in the blood.
A combination of a triaxial accelerometer and red and infrared PPG signals is used to process the signals through singular value decomposition (SVD), extract the AC and DC components of the signals, and calculate blood oxygen saturation by combining empirical coefficients.
It improves the accuracy and signal-to-noise ratio of blood oxygen saturation detection, and can more accurately reflect the binding of oxygen and hemoglobin in the blood.
Smart Images
Figure CN121817877A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of computer, in particular to a blood oxygen saturation detection method. BACKGROUND
[0002] Blood oxygen saturation, generally represented by S p O2, refers to the percentage of oxygen combined with hemoglobin in blood, and can reflect the function of respiratory and circulatory system. SUMMARY
[0003] In order to detect blood oxygen saturation, the present application provides a blood oxygen saturation detection method. The method comprises: obtaining a first signal; obtaining a second signal; obtaining a third signal by using the first signal and the second signal; obtaining a fourth signal by using the third signal; obtaining an alternating component of the fourth signal and a direct current component of the fourth signal by using the fourth signal; obtaining a fifth signal; obtaining a sixth signal by using the first signal and the fifth signal; obtaining a seventh signal by using the sixth signal; obtaining an alternating component of the seventh signal and a direct current component of the seventh signal by using the seventh signal; obtaining blood oxygen saturation.
[0004] The first signal can be an acceleration signal obtained by a three-axis acceleration sensor.
[0005] The second signal can be a red light PPG signal.
[0006] The second signal can also be a preprocessed red light PPG signal.
[0007] The third signal can be obtained by using the first signal and the second signal in the following way: e1(n)=d1(n)-y1(n); wherein e1(n) is the third signal; d1(n) is the second signal; y1(n) is obtained by the following formula: y1(n)=w1 T (n)x1(n); wherein x1(n) is the first signal; w1 T (n) is a transpose matrix of w1(n), and w1(n) is obtained by the following formula: w1(n)=w1(n-1)-μ1e1(n-1)x1(n-1); wherein μ1 is a first step parameter.
[0008] The third signal can also be obtained by using the first signal and the second signal in the following way: e1(n) = d1(n) - w2(n)x1(n); T (n)x1(n); wherein e1(n) is the third signal; d1(n) is the second signal; x1(n) is the first signal; w2(n) is a weight matrix of the second signal; and w2(n) is a transpose matrix of w2(n). T (n) is a transpose matrix of w2(n), and w2(n) is obtained by the following formula: w2(n) = w2(n-1) + 2μ2(n-1)x1(n-1)e1(n-1); wherein μ2(n-1) is obtained by the following formula: wherein x1(n-1) is the first signal; and x1(n-1) is a transpose matrix of x1(n-1). T
[0009] The third signal can also be obtained by using the first signal and the second signal in the following way: e1(n) = d1(n) - w3(n)x1(n); T (n)x1(n); wherein e1(n) is the third signal; d1(n) is the second signal; x1(n) is the first signal; w3(n) is a weight matrix of the second signal; and w3(n) is a transpose matrix of w3(n). T (n) is a transpose matrix of w3(n), and w3(n) is obtained by the following formula: w3(n) = w3(n-1) + k1(n)e1(n-1); wherein k1(n) is obtained by the following formula: wherein η1 is the first forgetting factor; x1(n) is the first signal; and x1(n) is a transpose matrix of x1(n). T (n) is a transpose matrix of x1(n); and q1(n-1) is obtained by the following formula: q1(n-1) = [q1(n-2) - k1(n-1)x1(n-1)q1(n-2)]. T
[0010] The third signal can also be obtained by using the first signal and the second signal in the following way: e1(n) = d1(n) - w4(n)x1(n); T (n)x1(n); wherein e1(n) is the third signal; d1(n) is the second signal; x1(n) is the first signal; w4(n) is a weight matrix of the second signal; and w4(n) is a transpose matrix of w4(n). T (n) is a transpose matrix of w4(n), and w4(n) is obtained by the following formula: w4(n) = w4(n-1) - μ4(n-1)e1(n-1)x1(n-1). wherein μ4(n-1) is obtained by μ4(n-1) = δ1μ4(n-2) + γ1p4 2 (n-2) ; wherein δ1 is a first step adjustment coefficient; γ1 is a second step adjustment coefficient; p4(n-2) is obtained by p4(n-2) = β1p4(n-3) + (1 - β1)e1(n-2)e1(n-3) ; wherein β1 is a third step adjustment coefficient.
[0011] The fourth signal can be obtained by using the third signal in the following way: The third signal is arranged into a first matrix H1: The first matrix is subjected to SVD decomposition to obtain H1 = U1Λ1V1 * , wherein H1 is a first matrix of L1×M1; U1 is a unitary matrix of L1×L1; V1 is a unitary matrix of M1×M1, V1 * is a conjugate transpose matrix of V1; Λ1 is a semi-positive definite diagonal matrix of L1×M1; The element with the first largest value and the element with the second largest value in matrix Λ1 are replaced by 0 to obtain Λ'1; U1Λ'1V1 * is subjected to SVD inverse transformation to obtain a second matrix H'1: H'1 = U1Λ'1V1 * ; The second matrix H'1 is converted into a one-dimensional signal, which is the fourth signal.
[0012] The fifth signal can be an infrared light PPG signal.
[0013] The fifth signal can also be a pre-processed infrared light PPG signal.
[0014] The sixth signal can be obtained by using the first signal and the fifth signal in the following way: e2(n) = d2(n) - y2(n) ; wherein e2(n) is the sixth signal; d2(n) is the fifth signal; y2(n) is obtained by y2(n) = w5 T (n) x 1(n) ; wherein x1(n) is the first signal; w5 T(n) is a transpose matrix of w5(n), and w5(n) is obtained by the following formula: w5(n) = w5(n - 1) - μ5e2(n - 1)x1(n - 1); wherein μ5 is a second step parameter.
[0015] The sixth signal can also be obtained by using the first signal and the fifth signal by the following method: e2(n) = d2(n) - w6 T (n)x1(n); wherein e2(n) is the sixth signal; d2(n) is the fifth signal; x1(n) is the first signal; w6 T (n) is a transpose matrix of w6(n), and w6(n) is obtained by the following formula: w6(n) = w6(n - 1) + 2μ6(n - 1)x1(n - 1)e2(n - 1); wherein μ6(n - 1) is obtained by the following formula: wherein x1 T (n - 1) is a transpose matrix of x1(n - 1).
[0016] The sixth signal can also be obtained by using the first signal and the fifth signal by the following method: e2(n) = d2(n) - w7 T (n)x1(n); wherein e2(n) is the sixth signal; d2(n) is the fifth signal; x1(n) is the first signal; w7 T (n) is a transpose matrix of w7(n), and w7(n) is obtained by the following formula: w7(n) = w7(n - 1) + k2(n)e2(n - 1); wherein k2(n) is obtained by the following formula: wherein η2 is a second forgetting factor; x1 T (n) is a transpose matrix of x1(n); q2(n - 1) is obtained by the following formula: q2(n - 1) = [q2(n - 2) - k2(n - 1)x1 T (n - 1)q2(n - 2)].
[0017] The sixth signal can also be obtained by using the first signal and the fifth signal by the following method: e2(n) = d2(n) - w8 T (n)x1(n); Wherein, e2(n) is the sixth signal; d2(n) is the fifth signal; x1(n) is the first signal; w8 T (n) is the transpose matrix of w8(n), and w8(n) is obtained by the following formula: w8(n) = w8(n-1) - μ8(n-1)e2(n-1)x1(n-1); Wherein, μ8(n-1) is obtained by the following formula: μ8(n-1) = δ2μ8(n-2) + γ2p8 2 (n-2); Wherein, δ2 is the fourth step adjustment coefficient; γ2 is the fifth step adjustment coefficient; p8(n-2) is obtained by the following formula: p8(n-2) = β2p8(n-3) + (1-β2)e2(n-2)e2(n-3); Wherein, β2 is the sixth step adjustment coefficient.
[0018] The seventh signal can be obtained by the following method: The sixth signal is arranged into a third matrix H2: The third matrix is subjected to SVD decomposition to obtain: H2 = U2Λ2V2 * , Wherein, H2 is a third matrix of L2×M2; U2 is a unitary matrix of L2×L2; V2 is a unitary matrix of M2×M2, V2 * is the conjugate transpose matrix of V2; Λ2 is a semi-positive definite diagonal matrix of L2×M2; The element with the first largest value and the element with the second largest value in the matrix Λ2 are replaced by 0 to obtain Λ′2; U2Λ′2V2 * is subjected to SVD inverse transformation to obtain a fourth matrix H′2: H′2 = U2Λ′2V2 * ; The fourth matrix H′2 is converted into a one-dimensional signal, and the one-dimensional signal is the seventh signal.
[0019] The blood oxygen saturation can be obtained by the following method: S P O2 = AR 2 + BR + C, Wherein, S P O2 is the blood oxygen saturation, A is the first empirical coefficient, B is the second empirical coefficient, C is the third empirical coefficient, and R is calculated by the following formula: Wherein, AC1 is the AC component of the fourth signal, DC1 is the DC component of the fourth signal, AC2 is the AC component of the seventh signal, and DC2 is the DC component of the seventh signal.
[0020] The blood oxygen saturation detection method proposed in this application can detect blood oxygen saturation. Attached Figure Description
[0021] Figure 1 This is a flowchart of the blood oxygen saturation detection method of this application. Detailed Implementation
[0022] The blood oxygen saturation detection method of this application is described in detail below with reference to the accompanying drawings.
[0023] This application provides a method for detecting blood oxygen saturation. For example... Figure 1 As shown, the method includes: Receive the first signal; Obtain the second signal; The third signal is obtained using the first and second signals; The fourth signal is obtained using the third signal; The AC component and DC component of the fourth signal are obtained using the fourth signal. Obtain the fifth signal; The sixth signal is obtained using the first and fifth signals; The seventh signal is obtained using the sixth signal; The AC component and DC component of the seventh signal are obtained using the seventh signal. Obtain blood oxygen saturation.
[0024] The first signal can be an acceleration signal obtained through a triaxial accelerometer. This signal can be obtained by having the person being tested wear a blood oxygen saturation detection device equipped with a triaxial accelerometer.
[0025] The second signal can be a red PPG (photoplethysmo-graphic) signal, which can be obtained by having the person being tested wear a blood oxygen saturation detection device.
[0026] Red light can be red light with a wavelength of 660nm.
[0027] Blood oxygen saturation detection devices can be wearable devices, such as fitness trackers.
[0028] The second signal can also be a pre-processed red PPG signal.
[0029] The third signal can be obtained using the first and second signals through the following method: e1(n) = d1(n) - y1(n); where e1(n) is a third signal; d1(n) is a second signal; y1(n) is obtained by y1(n) = w1 T (n)x1(n); where x1(n) is a first signal; w1 T (n) is a transpose matrix of w1(n), w1(n) is obtained by w1(n) = w1(n - 1) - μ1e1(n - 1)x1(n - 1); where μ1 is a first step parameter.
[0030] The third signal can also be obtained by using the first signal and the second signal by the following method: e1(n) = d1(n) - w2 T (n)x1(n); where e1(n) is a third signal; d1(n) is a second signal; x1(n) is a first signal; w2 T (n) is a transpose matrix of w2(n), w2(n) is obtained by w2(n) = w2(n - 1) + 2μ2(n - 1)x1(n - 1)e1(n - 1); where μ2(n - 1) is obtained by where x1 T (n - 1) is a transpose matrix of x1(n - 1).
[0031] The third signal can also be obtained by using the first signal and the second signal by the following method: e1(n) = d1(n) - w3 T (n)x1(n); where e1(n) is a third signal; d1(n) is a second signal; x1(n) is a first signal; w3 T (n) is a transpose matrix of w3(n), w3(n) is obtained by w3(n) = w3(n - 1) + k1(n)e1(n - 1); where k1(n) is obtained by where η1 is a first forgetting factor; x1 T (n) is a transpose matrix of x1(n); q1(n - 1) is obtained by q1(n-1)=[q1(n-2)-k1(n-1)x1 T (n-1)q1(n-2)].
[0032] The third signal can also be obtained using the first and second signals through the following method: e1(n)=d1(n)-w4 T (n)x1(n); Where e1(n) is the third signal; d1(n) is the second signal; x1(n) is the first signal; w4 T (n) is the transpose of w4(n), which is obtained by the following formula: w4(n)=w4(n-1)-μ4(n-1)e1(n-1)x1(n-1); Wherein, μ4(n-1) is obtained by the following formula: μ4(n-1)=δ1μ4(n-2)+γ1p4 2 (n-2); Where δ1 is the first step length adjustment coefficient; γ1 is the second step length adjustment coefficient; p4(n-2) is obtained by the following formula: p4(n-2)=β1p4(n-3)+(1-β1)e1(n-2)e1(n-3); Where β1 is the adjustment coefficient for the third step size.
[0033] The fourth signal can be obtained using the third signal through the following method: Arrange the third signal into the first matrix H1: Performing Singular Value Decomposition (SVD) on the first matrix yields: H1=U1Λ1V1 * , Where H1 is the first matrix of L1×M1; U1 is the unitary matrix of L1×L1; V1 is the unitary matrix of M1×M1, V1 * Λ1 is the conjugate transpose of V1; Λ1 is a positive semi-definite diagonal matrix of L1×M1; Replace the first and second largest elements in matrix Λ1 with 0 to obtain Λ′1. For U1Λ′1V1 * Performing the inverse SVD transformation, we obtain the second matrix H′1: H′1=U1Λ′1V1 * ; The second matrix H'1 is converted into a one-dimensional signal, which is the fourth signal.
[0034] The AC component of the fourth signal and the DC component of the fourth signal are obtained by using the fourth signal, which can be achieved by using the prior art, and thus will not be described herein.
[0035] The fifth signal can be an infrared light PPG (photoplethysmo-graphic) signal, which can be obtained by wearing a blood oxygen saturation detection device by the person to be detected.
[0036] The infrared light can be infrared light with a wavelength of 940 nm.
[0037] The blood oxygen saturation detection device can be a wearable blood oxygen saturation detection device, such as a sports bracelet.
[0038] The fifth signal can also be a pre-processed infrared light PPG signal.
[0039] The sixth signal can be obtained by using the first signal and the fifth signal by the following method: e2(n)=d2(n)-y2(n); wherein e2(n) is the sixth signal; d2(n) is the fifth signal; y2(n) is obtained by the following formula: y2(n)=w5 T (n)x1(n); wherein x1(n) is the first signal; w5 T (n) is a transpose matrix of w5(n), and w5(n) is obtained by the following formula: w5(n)=w5(n-1)-μ5e2(n-1)x1(n-1); wherein μ5 is a second step parameter.
[0040] The sixth signal can also be obtained by using the first signal and the fifth signal by the following method: e2(n)=d2(n)-w6 T (n)x1(n); wherein e2(n) is the sixth signal; d2(n) is the fifth signal; x1(n) is the first signal; w6 T (n) is a transpose matrix of w6(n), and w6(n) is obtained by the following formula: w6(n)=w6(n-1)+2μ6(n-1)x1(n-1)e2(n-1); wherein μ6(n-1) is obtained by the following formula: wherein x1 T(n-1) is a transpose matrix of x1(n-1).
[0041] The sixth signal can also be obtained by using the first signal and the fifth signal through the following method: e2(n) = d2(n) - w7 T (n)x1(n) ; wherein e2(n) is the sixth signal; d2(n) is the fifth signal; x1(n) is the first signal; w7 T (n) is a transpose matrix of w7(n), and w7(n) is obtained through the following formula: w7(n) = w7(n-1) + k2(n)e2(n-1) ; wherein k2(n) is obtained through the following formula: wherein η2 is a second forgetting factor; x1 T (n) is a transpose matrix of x1(n); q2(n-1) is obtained through the following formula: q2(n-1) = [q2(n-2) - k2(n-1)x1 T (n-1)q2(n-2)].
[0042] The sixth signal can also be obtained by using the first signal and the fifth signal through the following method: e2(n) = d2(n) - w8 T (n)x1(n) ; wherein e2(n) is the sixth signal; d2(n) is the fifth signal; x1(n) is the first signal; w8 T (n) is a transpose matrix of w8(n), and w8(n) is obtained through the following formula: w8(n) = w8(n-1) - μ8(n-1)e2(n-1)x1(n-1) ; wherein μ8(n-1) is obtained through the following formula: μ8(n-1) = δ2μ5(n-2) + γ2p8 2 (n-2) ; wherein δ2 is a fourth step adjustment coefficient; γ2 is a fifth step adjustment coefficient; p8(n-2) is obtained through the following formula: p8(n-2) = β2p8(n-3) + (1-β2)e2(n-2)e2(n-3) ; wherein β2 is a sixth step adjustment coefficient.
[0043] The seventh signal can be obtained by using the sixth signal through the following method: arranging the sixth signal into a third matrix 2: Performing Singular Value Decomposition (SVD) on the third matrix yields: H2=U2Λ2V2 * , Where H2 is the third matrix of L2×M2; U2 is the unitary matrix of L2×L2; V2 is the unitary matrix of M2×M2, V2 * Λ2 is the conjugate transpose of V2; Λ2 is a positive semi-definite diagonal matrix of L2×M2; Replace the first and second largest elements in matrix Λ2 with 0 to obtain Λ′2. For U2Λ′2V2 * Performing the inverse SVD transformation, we obtain the fourth matrix H′2: H′2=U2Λ′2V2 * ; The fourth matrix H′2 is converted into a one-dimensional signal, which is the seventh signal.
[0044] Obtaining the AC component and DC component of the seventh signal using the seventh signal can be achieved using existing technology, and will not be elaborated here.
[0045] Blood oxygen saturation can be obtained using the AC and DC components of the fourth signal and the AC and DC components of the seventh signal through the following formula: S P O2 = AR 2 +BR+C, Among them, S p O2 represents blood oxygen saturation, A is the first empirical coefficient, B is the second empirical coefficient, C is the third empirical coefficient, and R is calculated using the following formula: Wherein, AC1 is the AC component of the fourth signal, DC1 is the DC component of the fourth signal, AC2 is the AC component of the seventh signal, and DC2 is the DC component of the seventh signal.
[0046] In one embodiment of this application, the first empirical coefficient A is -22.537, the second empirical coefficient B is 1.4389, and the third empirical coefficient C is 103.78.
Claims
1. A method for detecting blood oxygen saturation, characterized in that, include: Receive the first signal; Obtain the second signal; The third signal is obtained using the first and second signals; The fourth signal is obtained using the third signal; The AC component and DC component of the fourth signal are obtained using the fourth signal. Obtain the fifth signal; The sixth signal is obtained using the first and fifth signals; The seventh signal is obtained using the sixth signal; The AC component and DC component of the seventh signal are obtained using the seventh signal. Obtain blood oxygen saturation.
2. The blood oxygen saturation detection method as described in claim 1, characterized in that, The method for obtaining the third signal using the first and second signals is as follows: e1(n) = d1(n) - y1(n); Where e1(n) is the third signal; d1(n) is the second signal; and y1(n) is obtained by the following formula: y1(n) = w1T(n)x1(n); Where x1(n) is the first signal; w1 T (n) is the transpose of w1(n), which is obtained by the following formula: w1(n)=w1(n-1)-μ1e1(n-1)x1(n-1); Where μ1 is the first step length parameter.
3. The blood oxygen saturation detection method as described in claim 1, characterized in that, The method for obtaining the third signal using the first and second signals is as follows: e1(n) = d1(n) - w2T(n)x1(n); Where e1(n) is the third signal; d1(n) is the second signal; x1(n) is the first signal; w2 T (n) is the transpose of w2(n), which is obtained by the following formula: w2(n)=w2(n-1)+2μ2(n-1)x1(n-1)e1(n-1); Wherein, μ2(n-1) is obtained by the following formula: Where x1 T (n-1) is the transpose of x1(n-1).
4. The blood oxygen saturation detection method as described in claim 1, characterized in that, The method for obtaining the third signal using the first and second signals is as follows: e1(n)=d1(n-1)-w3T(n)x1(n); Where e1(n) is the third signal; d1(n) is the second signal; x1(n) is the first signal; w3 T (n) is the transpose of w3(n), which is obtained by the following formula: w3(n)=w3(n-1)+k1(n)e1(n-1); Wherein, k1(n) is obtained by the following formula: Where η1 is the first forgetting factor; x1 T (n) is the transpose of x1(n); q1(n-1) is obtained by the following formula: q1(n-1)=[q1(n-2)-k1(n-1)x1 T (n-1)q1(n-2)]。 5. The blood oxygen saturation detection method as described in claim 1, characterized in that, The method for obtaining the third signal using the first and second signals is as follows: e1(n)=d1(n-1)-w4 T (n)x1(n); Where e1(n) is the third signal; d1(n) is the second signal; x1(n) is the first signal; w4 T (n) is the transpose of w4(n), which is obtained by the following formula: w4(n)=w4(n-1)-μ4(n-1)e1(n-1)x1(n-1); Wherein, μ4(n-1) is obtained by the following formula: μ4(n-1)=δ1μ4(n-2)+γ1p4 2 (n-2); Where δ1 is the first step length adjustment coefficient; γ1 is the second step length adjustment coefficient; p4(n-2) is obtained by the following formula: p4(n-2)=β1p4(n-3)+(1-β1)e1(n-2)e1(n-3); Where β1 is the adjustment coefficient for the third step size.
6. The method for detecting blood oxygen saturation as described in any one of claims 2-5, characterized in that, The method for obtaining the fourth signal using the third signal is as follows: Arrange the third signal into the first matrix H1: Performing SVD decomposition on the first matrix yields: H1=U1Λ1V1 * , Where H1 is the first matrix of L1×M1; U1 is the unitary matrix of L1×L1; V1 is the unitary matrix of M1×M1, V1 * Λ1 is the conjugate transpose of V1; Λ1 is a positive semi-definite diagonal matrix of L1×M1; Replace the first and second largest elements in matrix Λ1 with 0 to obtain Λ′1; For U1Λ′1V1 * Performing the inverse SVD transformation, we obtain the second matrix H′1: H′1=U1Λ′1V1 * ; The second matrix H′1 is converted into a one-dimensional signal, which is the fourth signal.
7. The blood oxygen saturation detection method as described in claim 1, characterized in that, The method for obtaining the sixth signal using the first and fifth signals is as follows: e2(n) = d2(n) - y2(n); Where e2(n) is the sixth signal; d2(n) is the fifth signal; and y2(n) is obtained by the following formula: y2(n)=w5 T (n)x1(n); Where x1(n) is the first signal; w5 T (n) is the transpose of w5(n), which is obtained by the following formula: w5(n)=w5(n-1)-μ5e2(n-1)x1(n-1); Where μ5 is the second step size parameter.
8. The blood oxygen saturation detection method as described in claim 1, characterized in that, The method for obtaining the sixth signal using the first and fifth signals is as follows: e2(n)=d2(n)-w6 T (n)x1(n); Where e2(n) is the sixth signal; d2(n) is the fifth signal; x1(n) is the first signal; w6 T (n) is the transpose of w6(n), which is obtained by the following formula: w6(n)=w6(n-1)+2μ6(n-1)x1(n-1)e2(n-1); Wherein, μ6(n-1) is obtained by the following formula: Where x1 T (n-1) is the transpose of x1(n-1).
9. The method for detecting blood oxygen saturation as described in any one of claims 7-8, characterized in that, The method for obtaining the seventh signal using the sixth signal is as follows: Arrange the sixth signal into the third matrix H2: Performing SVD decomposition on the third matrix yields: H2=U2Λ2V2 * , Where H2 is the third matrix of L2×M2; U2 is the unitary matrix of L2×L2; V2 is the unitary matrix of M2×M2, V2 * Λ2 is the conjugate transpose of V2; Λ2 is a positive semi-definite diagonal matrix of L2×M2; Replace the first and second largest elements in matrix Λ2 with 0 to obtain Λ′2; For U2Λ′2V2 * Performing the inverse SVD transformation, we obtain the fourth matrix H′2: H′2=U2Λ′2V2 * ; The fourth matrix H′2 is converted into a one-dimensional signal, which is the seventh signal.
10. The blood oxygen saturation detection method as described in claim 1, characterized in that, The blood oxygen saturation was obtained using the following method: S P O2=AR 2 +BR+C, Among them, S P O2 represents blood oxygen saturation, A is the first empirical coefficient, B is the second empirical coefficient, C is the third empirical coefficient, and R is calculated using the following formula: Wherein, AC1 is the AC component of the fourth signal, DC1 is the DC component of the fourth signal, AC2 is the AC component of the seventh signal, and DC2 is the DC component of the seventh signal.