Multi-wavelength optical rotation angle and dynamic spectrum blood glucose noninvasive measurement method of polarization camera

Abstract

The invention discloses a multi-wavelength optical rotation angle and dynamic spectrum blood glucose noninvasive measurement method of a polarization camera, and relates to the technical field of polarized light and spectral analysis, and the method comprises the following steps: taking the polarization camera as a sensor, and greatly improving the precision of the optical rotation angle and the dynamic spectrum through a method of superposing all pixel points in a collected image; a dynamic optical rotation angle and a dynamic spectrum are obtained by utilizing the idea of a dynamic spectrum theory, so that the influence caused by factors such as individual difference and instability of a light source in a measurement process is reduced, and the signal-to-noise ratio of data is improved; the dynamic optical rotation angle and the dynamic spectrum are combined for modeling, the information amount related to human body blood sugar components in the model building process is increased through the method of combining the two optical characteristics, and the modeling accuracy is improved. According to the multi-wavelength optical rotation angle and dynamic spectrum blood glucose noninvasive measurement method adopting the polarization camera, the problems that an existing detection method is low in signal to noise ratio and low in accuracy are solved, and high-precision measurement of noninvasive blood glucose content is achieved.

Description

Technical Field

[0001] The present invention relates to the field of measurement of optical rotation angle and absorbance value, and particularly to a non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectral blood glucose of a polarization camera. Background Art

[0002] The development of non-invasive quantitative analysis technology for human blood components not only contributes to applications in medicine, helping doctors diagnose diseases in real time, but most importantly, it can reduce the pain of patients during blood routine and blood biochemical tests. Especially for diabetic patients who need to frequently measure blood glucose, non-invasive methods will greatly reduce their pain of frequent fingertip blood sampling for measurement.

[0003] In the field of noninvasive quantitative analysis of human blood glucose, Chinese invention patent CN213758242U discloses a noninvasive blood glucose meter based on blood glucose optical rotation angle detection. The device features a housing, a polarized light transmitter, a polarized light receiver, and a control device. This device utilizes polarized light, which can penetrate human tissue, and the optical rotation effect of glucose in blood on the polarized light. The polarized light is then detected, and the resulting rotated polarized light angle is highly consistent with changes in blood glucose concentration, thereby enabling noninvasive blood glucose measurement. Chinese invention patent CN115736912A discloses a transmission-based noninvasive blood glucose measurement method and system. This system continuously collects ten photometric data points from red and infrared light passing through a finger. After the acquisition is complete, the maximum and minimum values with the largest fluctuations are removed, and the remaining data are superimposed and averaged. The measured photometric data is then input into a calculation model, which then uses this model to determine the blood glucose level. Chinese invention patent CN116965812A discloses a noninvasive blood glucose monitoring method and system based on fractional Fourier transform analysis. This method uses a PPG signal sensor to collect human blood flow signals. After preprocessing the blood flow signals, a fractional discrete Fourier transform is performed to extract the energy spectrum as data features. A random forest regression model is then trained on a dataset constructed based on fingertip blood glucose concentrations. After evaluating the model's prediction error and blood glucose accuracy, a blood glucose prediction model is constructed. Finally, the prediction model is used to calculate the fingertip blood glucose concentration of the target subject's blood flow signals. Chinese invention patent CN116491941A discloses a light-reflection-based noninvasive blood glucose monitoring device and system. This device utilizes multiple near-infrared light source modules to illuminate a person's finger. The signals are then detected using the same number of photodetectors as the near-infrared light source modules. The detected signals are then processed and analyzed. Chinese invention patent CN117179753A discloses a portable noninvasive blood glucose monitoring device and method based on multidimensional optical information. This method uses the incident light to illuminate the detection part of the human body and the reflected light formed by diffuse reflection is received by the detection module after passing through a series of detection windows. The measured Mueller matrix is then disassembled by the Mueller matrix polarization decomposition method to obtain 16-dimensional optical parameters. Finally, the effective optical parameters are obtained through the spatial metric learning dimensionality reduction method to construct the optical information metric space for neural network training.

[0004] While the above methods have achieved some success in noninvasively measuring blood glucose levels, the results are insufficient for clinical application. Current optical methods for noninvasive blood glucose measurement suffer from the following issues that affect accuracy: first, individual differences, such as skin texture, finger thickness, and tissue thickness; second, a low signal-to-noise ratio; and third, the complex composition of human tissue makes it difficult to achieve high-precision blood glucose measurements using a single optical method. The method proposed in this invention significantly reduces the impact of these factors. Summary of the Invention

[0005] To overcome the above problems, taking advantage of the optical rotation of blood glucose and based on Lambert-Beer's law, the present invention proposes to simultaneously obtain the optical rotation angles and dynamic spectra of multiple wavelengths using a polarization camera, and perform non-invasive detection of human blood glucose using the combined data of the two. The method includes the following steps: S1: A multi-channel adjustable constant current drive circuit generates constant current drive signals with preset different carrier frequencies, controls multiple light sources with different wavelengths using the above multiple constant current drive circuits, then combines the light sources with different wavelengths controlled by multiple different carrier frequencies into a beam of light using a multi-in-one optical fiber, and obtains linearly polarized light modulated by multiple wavelengths using this beam of light passing through a polarizer (linear polarizer); S2: When irradiating a human finger with linearly polarized light modulated by multiple wavelengths, elliptically polarized light of the human finger at multiple wavelengths is obtained; S3: Use a polarization camera to simultaneously collect consecutive multi-frame images of the light waves at four angles (0°, 45°, 90°, and 135°) of the elliptically polarized light; S4: Add up the gray values of all pixel points in each frame of the image obtained at each angle. The accumulated value is the light intensity value at this moment at this angle. The light intensity values obtained from all the images at each angle are arranged in the order of acquisition time to obtain digital signals at each angle within a period of time. The digital signals at the four angles are respectively , , and ; S5: Call the fast digital phase-locked demodulation algorithm to demodulate the digital signals at the four angles at each wavelength, obtain the digital signals at the four angles of multiple different wavelengths after demodulation ( , , and ), and based on the data at the four angles of all wavelengths after demodulation, use the Stokes vector method to calculate the dynamic optical rotation angle and transmission spectrum at each wavelength related to human blood respectively, and then use the dynamic spectrum extraction method to extract the dynamic spectrum at each wavelength; S6: Use a modeling method to establish a model between the combined feature information of the dynamic optical rotation angles and dynamic spectra at multiple wavelengths and the blood glucose content, and use the model to predict the blood glucose content of unknown samples.

[0006] Among them, the working mode of the multi-channel adjustable constant current drive circuit is: Drive the multi-channel adjustable constant current drive circuit according to signals with preset multiple different carrier frequencies, and then each light source controls the light source according to the working mode of the constant current drive module.

[0007] Among them, the generation method of the modulated linearly polarized light is: After passing through the polarizer, the multi-wavelength modulated polarized light becomes modulated linearly polarized light. A preset included angle is formed between the polarizer (polarizing plate) and the (polarization detection) direction of the polarization camera. The preset included angle needs to be set according to the actual experiment. In specific implementation, the embodiments of the present invention do not limit this.

[0008] Among them, the polarization camera specifically works as follows: There are multiple CMOS optoelectronic conversion sensors on the chip of the polarization camera, and four polarizing plates at different angles (0°, 45°, 90°, and 135°) are respectively placed on each sensor. When the polarization camera works, at a preset frame rate ( ) collect frame images, where each frame image contains multiple pixel points at four different angles. Connect and arrange the pixels corresponding to all the same-named polarizing plates in each frame image collected in the order from left to right and from top to bottom to obtain images at each angle.

[0009] Among them, the method for setting the size of the preset frame rate is as follows: The preset frame rate ( ) is at least times the maximum carrier frequency, where is a constant.

[0010] Among them, the fast digital lock-in demodulation algorithm specifically is: When demodulating the modulation signals at each wavelength, the digital signals at four angles (0°, 45°, 90°, and 135°) collected , , and , are respectively downsampled to 4 times the carrier frequency ( ) corresponding to the wavelength, and then divided into 4 groups in the order of sampling. Each group data points, and then perform a summation operation on all the data points in each group to obtain the downsampled digital signals , , and at each wavelength. The fast digital lock-in algorithm realizes the discrimination of signal amplitude and phase by multiplying the collected signal by two orthogonal reference signals. One-period sine and cosine reference signal sequences are respectively: , (1) Let four consecutive points be a period, and the cross-correlation signal after low-pass filtering corresponding to the signal of each period is: (2) (3) Then the amplitudes and phases of the signals demodulated in each period are calculated as (4) (5) According to the above addition and subtraction operations, digital signals at four angles at each wavelength after demodulation are obtained 、 、 and .

[0011] Among them, the calculation process of using Stokes vector parameters to calculate the dynamic optical rotation angles of human blood at each wavelength is as follows: Calculate the value at each wavelength, then the angle between the linearly polarized light containing human blood information and the 0° plane at each wavelength is: (6) Then calculate the maximum value and the minimum value in each pulse wave period of the signal at each wavelength, and then subtract the two values: (7) The cumulative value of the values of all pulse wave periods in the signal at each wavelength is the dynamic optical rotation angle of human blood at the corresponding wavelength.

[0012] Among them, the calculation process of using Stokes vector parameters to calculate the transmission spectra of human blood at multiple wavelengths is as follows: Calculate the value at each wavelength: , is the transmission spectrum of human blood at the corresponding wavelength.

[0013] Among them, the calculation process of using the dynamic spectrum extraction method to calculate the dynamic spectra of human blood at multiple wavelengths is as follows: Perform a logarithmic operation on the signal: (8) Then use the dynamic spectrum extraction method to extract the dynamic spectrum of human blood in the signal at each wavelength, and obtain the absorbance values of the components in human blood.

[0014] Among them, the specific operation of the modeling method is as follows: Establish the relationship between the characteristic parameters and the blood glucose content.

[0015] The beneficial effects of the technical solution provided by the present invention are as follows: The method provided by the present invention makes full use of the collected data, improves the signal-to-noise ratio of the data, reduces the influence caused by individual differences, increases the information content of different optical characteristics of human blood glucose components, and thus improves the accuracy of non-invasive detection of human blood glucose components. Description of the Drawings

[0016] Figure 1 It is a flowchart of the non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectrum blood glucose of the polarization camera provided by the embodiment of the present invention.

[0017] Figure 2 It is an experimental device diagram of the non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectrum blood glucose of the polarization camera provided by the embodiment of the present invention.

[0018] Figure 3 It is a flowchart of obtaining multi-wavelength modulated linearly polarized light provided by the embodiment of the present invention.

[0019] Figure 4 It is a flowchart of obtaining optical wave signals of four angles of modulated elliptically polarized light containing human finger information at multiple wavelengths provided by the embodiment of the present invention.

[0020] Figure 5 It is a flowchart of calculating the dynamic spectrum and dynamic optical rotation angle of human blood provided by the embodiment of the present invention.

[0021] Figure 6 It is a flowchart of modeling and quantitatively analyzing human blood glucose provided by the embodiment of the present invention. Detailed Embodiments

[0022] To make the objectives, technical solutions, and advantages of the present invention clearer, the following further describes the embodiments of the present invention in detail.

[0023] In order to further reduce the errors caused by individual differences and environmental interference, improve the signal-to-noise ratio of the characteristic information, increase the information content related to the blood glucose content, and obtain more accurate prediction results, the embodiment of the present invention provides a non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectrum blood glucose of a polarization camera. Refer to Figure 1 and Figure 2 , and the details are described below.

[0024] 101: The multi-channel adjustable constant current drive circuit generates constant current drive signals with different carrier frequencies, uses the constant current drive circuit to control multiple light sources with different wavelengths, then uses a multi-in-one optical fiber to combine the light sources with different wavelengths controlled by multiple different carrier frequencies into a single beam of light, and uses this single beam of light to pass through a polarizer (linear polarizer) to obtain linearly polarized light modulated by multiple wavelengths. When using the linearly polarized light modulated by multiple different wavelengths to irradiate the object to be measured, multi-wavelength polarized elliptical light containing human finger information is obtained;

[0025] This step specifically includes steps 1011 - 1014, see Figure 3 , as described in detail below.

[0026] 1011: Multiple adjustable constant current drive circuits generate constant current drive signals with a preset multiple of different carrier frequencies;

[0027] Among them, the constant current drive signal can be a square wave or other waveforms. In specific implementation, the embodiments of the present invention do not limit this;

[0028] Among them, the multiple different frequencies can be arranged in a geometric progression or set in other ways. In specific implementation, the embodiments of the present invention do not limit this;

[0029] 1012: Multiple constant current circuits with different frequencies drive multiple light sources with different wavelengths to generate multi-wavelength modulated polarized light signals;

[0030] Among them, the light source can be a laser or other types of polarized light sources. The embodiments of the present invention do not limit this;

[0031] Among them, the number of the multiple different wavelengths can be any number, and the wavelengths can be any different wavelengths. The embodiments of the present invention do not limit this;

[0032] 1013: The multi-wavelength modulated light signal becomes a multi-wavelength modulated linearly polarized light signal after passing through the polarizer;

[0033] Among them, after the multi-wavelength modulated light passes through the linear polarizer, the light with a polarization direction different from that of the linear polarizer will be absorbed or reflected, and the light with a polarization direction the same as that of the linear polarizer will pass through the linear polarizer, making the light passing through the linear polarizer become modulated linearly polarized light;

[0034] Among them, a preset angle is formed between the polarizer (polarizing sheet) and the (polarization detection) direction of the polarization light camera. The preset angle needs to be set according to actual experiments. In specific implementation, the embodiments of the present invention do not limit this.

[0035] 1014: The multi-wavelength modulated linearly polarized light signal is transmitted through a human finger to obtain multi-wavelength adjustable elliptically polarized light containing human finger information;

[0036] Among them, the human finger contains a solution with optical activity; a multi-wavelength modulated linearly polarized light is vertically incident on the object to be measured. After passing through the human finger, the polarization direction of the multi-wavelength modulated linearly polarized light will change. The magnitude and direction of the change are related to the human finger and the concentration of the optically active substances in the blood within the finger, and a multi-wavelength tunable elliptically polarized light containing human finger information is obtained.

[0037] 102: Use a polarization camera to simultaneously collect consecutive multi-frame images of the light waves at four angles (0°, 45°, 90°, and 135°) of the multi-wavelength tunable elliptically polarized light. Add up the gray values of all pixel points in each frame of the image obtained at each angle. The accumulated value is the light intensity value at this moment for this angle. The light intensity values obtained from the images at all moments under each angle are arranged in the order of the acquisition time to obtain digital signals for each angle within a period of time. The digital signals for the four angles are respectively 、 、 and ;

[0038] This step specifically includes steps 1021 - 1023, see Figure 4 , as described in detail below.

[0039] 1021: The polarization camera uses linear polarizers at four different angles (0°, 45°, 90°, and 135°) to decompose the multi-wavelength polarized elliptical light containing human finger information into light waves at four angles. The incident surface of the polarized elliptical light on the linear polarizer has an internal structure that can selectively transmit light waves vibrating in a specific direction. Only light in a specific direction can pass through the polarizer. Thus, after the multi-wavelength polarized elliptical light passes through four polarizers at different angles, light waves at four angles are respectively obtained;

[0040] Among them, the linear polarization beam splitter device can be linear polarizers at different angles or other devices;

[0041] 1022: There are multiple CMOS photoelectric conversion sensors on the chip of the polarization camera, and linear polarizers at four different angles (0°, 45°, 90°, and 135°) are respectively placed on each sensor. When the polarization camera works, it acquires frames of images at a preset frame rate ( ). Each frame of the image contains multiple pixel points at four different angles. Connect and arrange the pixels corresponding to the same-named polarizers in each frame of the acquired image in sequence from left to right and from top to bottom to obtain images for each angle;

[0042] 1023: Add up the gray values of all pixel points in the images at each angle at each moment. This accumulated value is the light intensity value at each moment for each angle. Arrange them in chronological order to obtain digital signals for the four angles , , and .

[0043] 103: Call the fast digital phase-locked demodulation algorithm to demodulate the signal, the signal, the signal and the signal respectively, and obtain the human finger information of each wavelength contained in the demodulated signal, the human finger information of each wavelength contained in the demodulated signal, the human finger information of each wavelength contained in the demodulated signal, the human finger information of each wavelength contained in the demodulated signal, and the human finger information of each wavelength contained in the demodulated signal, and based on the of each wavelength after demodulation, the of each wavelength after demodulation, the of each wavelength after demodulation, the of each wavelength after demodulation, the of each wavelength after demodulation, and the of each wavelength after demodulation, use the Stokes vector method to calculate the dynamic spectrum and dynamic optical rotation angle of human blood at each wavelength.

[0044] This step specifically includes steps 1031 - 1035, see Figure 5 , details are described below.

[0045] 1031: Use the fast digital phase-locked demodulation algorithm to demodulate the digital signals , , and , and obtain the information of the human finger contained in the four directions of the elliptically polarized light of each wavelength respectively: , , and ;

[0046] Among them, use formulas (1), (2), (3), (4) and (5) for demodulation operations to obtain the light intensity value information of the four angles at each wavelength.

[0047] 1032: Use the Stokes vector method to obtain the signal of the angle between the linearly polarized light containing human finger information and the 0° plane at each wavelength ;

[0048] Among them, calculate the optical rotation angle at each moment at each wavelength according to formula (6).

[0049] 1033: Search for the maximum value of each pulse wave cycle in the signal at each wavelength and the minimum value , and finally subtract the two values to obtain only the optical rotation angle of human blood: , and finally for each wavelength average the optical rotation angles of all cycles in the signal, and the average value is the dynamic optical rotation angle of human blood at each wavelength;

[0050] Among them, the optical rotation angle of human blood is positively correlated with the concentration of optically active substances in the blood. By calculating the optical rotation angle of human blood, the concentration of blood sugar can be indirectly obtained. During the measurement process, individual differences, the light source and the human body will inevitably fluctuate, which will affect , , and the measurement results, and the dynamic optical rotation angle can effectively reduce the influence caused by the above factors.

[0051] 1034: Calculate the value at each wavelength : , and obtain the transmitted light intensity value information containing human finger information at each wavelength;

[0052] 1035: Take the logarithm of the signal at each wavelength : , and then use the dynamic spectrum extraction method to extract the dynamic spectrum of human blood from the data at each wavelength .

[0053] Among them, the absorbance value of human blood is positively correlated with the concentration of components in the blood. By calculating the absorbance value of human blood, the concentration of human blood sugar can be indirectly obtained. During the measurement process, individual differences, the light source and the human body will inevitably fluctuate, which will affect and the measurement results, and the dynamic spectrum extraction method can effectively reduce the influence caused by the above factors.

[0054] Among them, the dynamic spectrum extraction method can be the difference extraction method, the single-edge extraction method or other dynamic spectrum extraction methods, and the embodiments of the present invention do not limit this.

[0055] 104: Use the modeling method to establish a model between the joint characteristic information of the dynamic optical rotation angle and the dynamic spectrum at multiple wavelengths and the human blood sugar content, and use the model to predict the human blood sugar content.

[0056] This step specifically includes steps 1041-1043, see Figure 6 , and the details are described below.

[0057] 1041: Using the modeling set and prediction set selection method, all samples are divided into a modeling set and a prediction set;

[0058] Among them: The modeling set and prediction set selection method can be an equal-interval extraction method or other methods, and the embodiments of the present invention do not limit this;

[0059] 1042: Using a modeling method to establish a model between the characteristic information of the dynamic optical rotation angle and the dynamic spectrum at all wavelengths of the samples in the modeling set and the blood glucose content: , obtaining a coefficient matrix ;

[0060] Among them: The modeling method can be the partial least squares method or other methods, and the embodiments of the present invention do not limit this;

[0061] Among them: is the blood glucose content, is the characteristic coefficient matrix of the model.

[0062] 1043: Using the said coefficient matrix can calculate the predicted value of the blood glucose content of each sample in the prediction set. Multiply the characteristic information of each sample in the prediction set by the corresponding coefficient in the coefficient matrix , and then the cumulative value after multiplying all the characteristic information by the corresponding coefficient matrix is the predicted value of the blood glucose content of the sample.

[0063] The constant current drive circuit and the like applied in the method of the embodiments of the present invention are all well-known technologies in data processing methods and are well-known to those skilled in the art.

[0064] In summary, the embodiments of the present invention provide a non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectrum of blood glucose of a polarization camera. Compared with the prior art, this method uses multiple constant current drive circuits to modulate light sources of multiple different wavelengths by means of frequency division multiplexing, eliminating mechanical structures such as choppers required by traditional technologies, and at the same time increasing the information volume of the target component at multiple wavelengths in the measurement system; using a polarization camera to simultaneously collect image data of four angles of elliptically polarized light, and taking the cumulative value obtained by adding all pixel points in each frame of image at each angle as the light intensity value at each angle at each moment, and using the method of converting space into precision to improve the precision and sensitivity of the data; the extraction method of dynamic optical rotation angle and dynamic spectrum reduces the influence of factors such as individual differences and light source instability in the measurement process on the optical rotation angle and absorbance value from the measurement principle, and improves the signal-to-noise ratio of the optical rotation angle and dynamic spectrum; the method of jointly establishing a model using dynamic spectrum and dynamic optical rotation angle increases the information volume of various characteristics related to the target component in the modeling process and improves the accuracy of modeling.

[0065] Those skilled in the art can understand that the attached drawings are only schematic diagrams of a preferred embodiment. The serial numbers of the above embodiments of the present invention are only for description and do not represent the superiority or inferiority of the embodiments. The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. Non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectral blood glucose of a polarization camera, characterized in that, The method includes the following steps: S1: A multi-channel adjustable constant current drive circuit generates constant current drive signals with preset different carrier frequencies, controls light sources with multiple different wavelengths using the above-mentioned multiple constant current drive circuits, then combines the light sources with different wavelengths controlled by multiple different carrier frequencies into a beam of light using a multi-in-one optical fiber, and obtains linearly polarized light modulated by multiple wavelengths by passing this beam of light through a polarizer (linear polarizer); S2: When irradiating a human finger with the linearly polarized light modulated by multiple wavelengths, elliptically polarized light of the human finger at multiple wavelengths is obtained; S3: A polarization camera simultaneously acquires consecutive multi-frame images of light waves at four angles (0°, 45°, 90°, and 135°) of the elliptically polarized light; S4: Accumulate the grayscale values of all pixel points in each frame of the image at each obtained angle. The accumulated value is the light intensity value at this moment for this angle. The light intensity values obtained from all images at each angle are arranged in the order of acquisition time to obtain the digital signals for each angle within a period of time. The digital signals for the four angles are respectively , , and ; S5: Call the fast digital phase-locked demodulation algorithm to demodulate the digital signals at the four angles under each wavelength respectively, and obtain the digital signals at the four angles of multiple different wavelengths after demodulation ( , , and ). Then, based on the data of the four angles at all wavelengths after demodulation, use the Stokes vector method to calculate the dynamic optical rotation angle and transmission spectrum at each wavelength of human blood respectively, and then use the dynamic spectrum extraction method to extract the dynamic spectrum at each wavelength; S6: A model between the joint characteristic information of the dynamic rotation angle and the dynamic spectrum at multiple wavelengths and the blood glucose content is established using a modeling method, and the blood glucose content of an unknown sample is predicted using the model.

2. The non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectrum blood glucose of the polarization camera according to claim 1, wherein The generation method of the frequency division modulated multi-wavelength light is as follows: According to signals with preset multiple different carrier frequencies, a multi-channel adjustable constant current drive circuit is driven, and then the light source of each wavelength controls the light source according to the working mode of the constant current drive module. The frequency division multiplexing method and coding method of the signals with preset multiple different carrier frequencies can be in a geometric progression manner or other methods, and the embodiments of the present invention do not limit this.

3. The non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectrum blood glucose of the polarization camera according to claim 1, characterized in that, The generation method of the multi-wavelength modulated linearly polarized light is as follows: The multi-wavelength modulated polarized light becomes multi-wavelength modulated linearly polarized light after passing through a polarizer (linear polarizer), where the angle between the polarizer (polarizer) and the (polarization detection) direction of the polarization camera is a preset angle, and the preset angle needs to be set according to actual experiments, and the embodiments of the present invention do not limit this.

4. The non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectrum blood glucose of the polarization camera according to claim 1, characterized in that, The specific operation of the polarization camera is as follows: There are multiple CMOS optoelectronic conversion sensors on the chip of the polarization camera, and four polarizers at different angles (0°, 45°, 90°, and 135°) are respectively placed on each sensor. When the polarization camera works, at a preset frame rate ( ) capture frame images, where each frame image contains multiple pixel points at four different angles. Connect and arrange the pixels corresponding to all the same-named polarizers in each captured frame image from left to right and from top to bottom to obtain images for each angle. Among them, the preset frame rate ( ) is at least times the maximum carrier frequency, where is a constant.

5. The non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectrum blood glucose of the polarization camera according to claim 1, wherein The specific fast digital lock-in demodulation algorithm is as follows: When demodulating the modulation signals at each wavelength, the digital signals at the four acquired angles (0°, 45°, 90°, and 135°) , , and are respectively downsampled to 4 times the carrier frequency ( ) corresponding to the wavelength, and then divided into 4 groups in the order of sampling. Each group data points, and then the sum operation is performed on all the data points in each group to obtain the downsampled digital signals at each wavelength , , and . The fast digital phase-locked algorithm realizes the discrimination of signal amplitude and phase by multiplying the acquired signal with two orthogonal reference signals (the positive and cosine reference signal sequences of one period are respectively: , ): Let four consecutive points be one period, and the cross-correlation signal after low-pass filtering corresponding to the signal of each period is: and ， The amplitudes and phases of the signals demodulated in each period are calculated as follows: and , and through the operations in each period, digital signals at four angles under each wavelength after demodulation are obtained , , and .

6. The non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectrum blood glucose of the polarization camera according to claim 1, characterized in that, The specific calculation process of calculating the dynamic rotation angle of human blood at each wavelength using the Stokes vector method is as follows: Calculate the value at each wavelength. Then, the angle between the linearly polarized light containing human finger information and the 0° plane at each wavelength is: . Then, calculate the maximum value and the minimum value in each pulse wave period of the signal at each wavelength. Then subtract the two values: . The cumulative value of the values of all pulse wave periods in the signal at each wavelength is the dynamic optical rotation angle of human blood at the corresponding wavelength.

7. The non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectrum blood glucose of the polarization camera according to claim 1, characterized in that, The specific calculation process of calculating the transmission spectrum of human blood at each wavelength using the Stokes vector method is as follows: Calculate the value at each wavelength: , which is the transmission spectrum of human blood at the corresponding wavelength.

8. The non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectrum blood glucose of the polarization camera according to claim 1, wherein The specific calculation process of calculating the dynamic spectrum of human blood at each wavelength using the dynamic spectrum extraction method is as follows: Pair Signal logarithmic operation: , and then use the dynamic spectrum extraction method to calculate the dynamic spectrum of human blood in the signal at each wavelength . Among them, the dynamic spectrum extraction method can be the difference extraction method, the single-edge extraction method or other dynamic spectrum extraction methods, and the embodiments of the present invention are not limited thereto.

9. The non-invasive measurement method for multi-wavelength optical rotation angle and dynamic spectrum blood glucose of the polarization camera according to claim 1, wherein The specific calculation process of establishing a model between the joint characteristic information of the dynamic spectrum and the dynamic rotation angle at multiple wavelengths and the blood glucose content using a modeling method is as follows: Using the method for selecting the modeling set and the prediction set, all samples are divided into a modeling set and a prediction set; using a modeling method to establish a model between the characteristic information of the dynamic optical rotation angle and the dynamic spectrum at all wavelengths of the samples in the modeling set and the blood glucose content: , obtaining a coefficient matrix ; using the coefficient matrix can calculate the predicted value of the blood glucose content of each sample in the prediction set. Multiply the characteristic information of each sample in the prediction set by the corresponding coefficient in the coefficient matrix , and then the cumulative value of all the characteristic information multiplied by the corresponding coefficient matrix is the predicted value of the blood glucose content of the sample. Among them: The method for selecting the modeling set and the prediction set can be an equal-interval sampling method or other methods, and the embodiments of the present invention do not limit this. Among them; the modeling method can be the partial least squares method or other modeling methods, and the embodiments of the present invention do not limit this.

Images