OCTA-based non-invasive continuous blood glucose measurement method and apparatus

Abstract

An OCTA-based non-invasive continuous blood glucose measurement method and apparatus. The OCTA-based non-invasive continuous blood glucose measurement method comprises: performing continuous OCT scanning imaging on a target tissue region, then using a depth attenuation characteristic of an OCT signal to extract a three-dimensional scattering coefficient of the target tissue region, and generating a three-dimensional microvascular distribution of the target tissue region; then accurately distinguishing a blood scattering coefficient and a tissue scattering coefficient on the basis of the three-dimensional scattering coefficient and the three-dimensional microvascular distribution; using an electrochemical method to measure reference blood glucose concentrations at two moments, and finally, respectively combining the reference blood glucose concentrations with the blood scattering coefficient and the tissue scattering coefficient to achieve continuous measurement of a blood glucose concentration and an interstitial fluid glucose concentration. According to the method, on the basis of OCTA guidance, the synchronous measurement of the interstitial fluid glucose concentration is implemented while accurate measurement of the blood glucose concentration is implemented, and the method has a great heuristic significance for the optical non-invasive blood glucose measurement.

Description

A non-invasive continuous blood glucose measurement method and device based on OCTA Technical Field

[0001] The present invention belongs to the field of biomedical testing and relates to a blood glucose measurement method and device based on optical coherence tomography angiography (OCTA), and in particular to a non-invasive continuous blood glucose measurement method and device based on OCTA. Technical Background

[0002] Diabetes is a chronic metabolic disease that can lead to serious complications such as heart disease, kidney problems, and stroke. Monitoring blood glucose levels is crucial to a patient's overall health. Glucose levels are typically assessed in two areas: the blood and interstitial fluid. Blood glucose concentrations are typically measured using enzyme-based electrochemical devices that require a blood sample to be collected through a painful finger or forearm prick, which can lead to discomfort, infection risks, and poor compliance. Another method for continuously monitoring interstitial fluid glucose concentrations is using a subcutaneous needle sensor. However, this method is less accurate than blood glucose monitoring due to a delayed response of interstitial fluid glucose levels and potential skin irritation.

[0003] Optical techniques that exploit near-infrared light scattering hold great potential for noninvasive glucose monitoring. However, detecting the subtle scattering changes caused by glucose against a strong scattering background is a significant obstacle for glucose monitoring. While optical coherence tomography can identify glucose-induced scattering changes in specific tissue layers, distinguishing between blood and interstitial fluid remains a significant obstacle. Blood and interstitial fluid glucose have opposite effects on the optical scattering properties of surrounding tissues, significantly impacting the accuracy of glucose measurements.

[0004] Optical coherence tomography angiography (OCTA) can generate three-dimensional vascular maps of blood perfusion at the capillary level. However, how to use OCTA to distinguish between blood and interstitial fluid and how to measure blood glucose after differentiation is a challenge for those skilled in the art.

[0005] Summary of the Invention

[0006] To address the challenges presented by the prior art, the present invention aims to provide a non-invasive, continuous blood glucose measurement method and device based on OCTA. The OCTA vascular mask of the present invention distinguishes between the blood and tissue scattering coefficients, enabling high-precision measurement of glucose-induced scattering changes in blood or interstitial fluid. Based on the linear relationship between the scattering coefficient and glucose concentration, OCTA technology has the potential to achieve high-precision, non-invasive, and continuous blood glucose measurement.

[0007] The technical solution adopted in the present invention is:

[0008] 1. A non-invasive continuous blood glucose measurement method based on OCTA

[0009] An OCTA continuous imaging method: performing continuous OCT scanning imaging on the target tissue area to obtain OCT signals;

[0010] A method for calculating the scattering coefficients of blood and surrounding tissues: generating a three-dimensional scattering coefficient and a three-dimensional microvascular distribution of a target tissue region based on OCT signals, and separately calculating and obtaining the blood scattering coefficient and the tissue scattering coefficient based on the three-dimensional scattering coefficient and the three-dimensional microvascular distribution;

[0011] A method for calibrating and calculating blood glucose concentration: blood samples are collected and reference blood glucose concentrations at two times, t1 and t2, are measured using an electrochemical method. The reference blood glucose concentrations and blood scattering coefficients at the same time are paired and then linearly fitted to obtain the blood glucose concentration (BGC) at all times.

[0012] A method for calibrating and calculating interstitial fluid glucose concentration: data pairing is performed on the reference blood glucose concentration at time points t1 and t2 and the tissue scattering coefficient at the corresponding delay time ΔT, and then linear fitting is performed to obtain the interstitial fluid glucose concentration IGC at all times.

[0013] The OCTA continuous imaging method is one of the following methods:

[0014] Time domain OCT imaging method by changing the optical path length of the reference arm through scanning;

[0015] Alternatively, a spectral domain OCT imaging method using a spectrometer to record spectral interference signals;

[0016] Alternatively, a swept-frequency OCT imaging method is used to record spectral interference signals using a swept-frequency light source.

[0017] The method for calculating the scattering coefficient of blood and surrounding tissues specifically includes:

[0018] Utilizing the depth attenuation characteristics of OCT, the three-dimensional scattering coefficient of the target tissue area in the OCT signal is extracted;

[0019] Generate three-dimensional microvascular distribution in the target tissue area based on OCT signals;

[0020] After the three-dimensional microvascular distribution image is binarized, the blood vessel mask and tissue mask are obtained. The blood scattering coefficient and tissue scattering coefficient are obtained by performing dot multiplication operation on the three-dimensional scattering coefficient with the blood vessel mask and tissue mask respectively.

[0021] The method of utilizing the depth attenuation characteristics of OCT to extract the three-dimensional scattering coefficient of the target tissue area in the OCT signal specifically includes:

[0022] First, the system noise of the OCT signal in the depth direction is removed to obtain the denoised OCT signal;

[0023] Then, the attenuation of the denoised OCT signal along the depth direction is compensated using a Gaussian fitting model to obtain a compensated OCT signal;

[0024] Finally, the scattering coefficient of the target tissue area is calculated based on the variation characteristics of the compensated OCT signal in the depth direction.

[0025] Generating the three-dimensional microvascular distribution of the target tissue area based on the OCT signal specifically includes: using the OCT blood flow signal extraction method to analyze the amplitude or phase or the amplitude and phase of the OCT signal to obtain the three-dimensional microvascular distribution of the target tissue area.

[0026] The OCT blood flow signal extraction method includes: differential calculation, speckle variance calculation, decorrelation calculation or eigendecomposition calculation.

[0027] After the three-dimensional microvascular distribution image is binarized, a blood vessel mask and a tissue mask are obtained, and the three-dimensional scattering coefficient is respectively multiplied by the blood vessel mask and the tissue mask to obtain the blood scattering coefficient and the tissue scattering coefficient, which specifically includes:

[0028] The pixel values ​​of all blood flow areas in the three-dimensional microvascular distribution are set to 1, and the pixel values ​​of non-blood flow areas are set to 0 to obtain the blood vessel mask. The blood flow scattering coefficient is extracted by performing a dot product operation on the blood vessel mask and the three-dimensional scattering coefficient.

[0029] The pixel values ​​of all blood flow areas in the three-dimensional microvascular distribution are set to 0, and the pixel values ​​of the non-blood flow areas are set to 1 to obtain the tissue mask. The tissue mask is then multiplied with the three-dimensional scattering coefficient to extract the tissue scattering coefficient.

[0030] 2. A non-invasive continuous blood glucose measurement device based on OCTA

[0031] an optical coherence tomography device for performing continuous OCT imaging of a target area;

[0032] One or more processors for dynamic real-time processing of data.

[0033] The optical coherence tomography device is one of the following:

[0034] Includes low-coherence light source, interferometer and detector;

[0035] or include a low-coherence light source, an interferometer, and a spectrometer;

[0036] or includes a swept wide spectrum light source, an interferometer, and a detector;

[0037] Or other devices capable of achieving optical coherence tomography.

[0038] The beneficial effects of the present invention are:

[0039] This invention, guided by OCTA, not only accurately measures blood glucose concentration but also simultaneously measures interstitial fluid glucose concentration. Compared to existing continuous, noninvasive blood glucose measurement technologies, this invention offers higher accuracy and better sensitivity, and has significant implications for continuous, noninvasive optical blood glucose measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] FIG1 is a schematic diagram of the method of the present invention;

[0041] FIG2 is a schematic diagram of the device of the present invention;

[0042] FIG3 is a schematic diagram of an embodiment of the present invention;

[0043] FIG4 is a schematic diagram of the blood glucose detection process of the present invention;

[0044] FIG5 shows a three-dimensional OCT structure, a three-dimensional scattering coefficient, and a three-dimensional OCTA angiogram according to an embodiment of the present invention;

[0045] FIG6 is a typical OCTA image according to an embodiment of the present invention;

[0046] 7 is a graph showing the relationship between the optical blood glucose concentration, optical interstitial fluid glucose concentration, and reference blood glucose concentration obtained using the method of the present invention during an oral glucose tolerance test (OGTT) according to an embodiment of the present invention and changes over time.

[0047] Among them: 1- an OCTA continuous imaging method; 2- a method for calculating the scattering coefficient of blood and surrounding tissues; 3- a method for calibrating and calculating blood glucose concentration; 4- a method for calibrating and calculating interstitial fluid glucose concentration; 11- a light source; 12- a beam splitter; 13- a reference arm collimator; 14- a plane high reflective mirror; 15- a sample arm collimator; 16- a scanning galvanometer; 17- an objective lens; 18- a sample to be measured; 19- an interference signal detection device; 20- a signal processor; 21- extracting the three-dimensional scattering coefficient of the target tissue region from the OCT signal using the depth attenuation characteristics of OCT; 22- generating the three-dimensional microvascular distribution of the target tissue region based on the OCT signal; 23- After the three-dimensional microvascular distribution image is binarized, the blood vessel mask and the tissue mask are obtained. After the three-dimensional scattering coefficient is dot-multiplied with the blood vessel mask and the tissue mask respectively, the blood scattering coefficient and the tissue scattering coefficient are obtained; 24-polarization controller; 31-broadband swept light source; 32-fiber coupler with a splitting ratio of 80:20; 33-fiber coupler with a splitting ratio of 50:50; 34-first optical circulator; 35-second optical circulator; 36-polarization controller; 40-first collimator; second collimator 37; 38-focusing lens; 42-scanning lens; 39-flat high-reflection mirror; 41-scanning galvanometer; 43-sample; 44-balanced detector; 45-processor. DETAILED DESCRIPTION

[0048] The present invention will be further described in detail below with reference to the accompanying drawings, which form a part of this document. It should be noted that these descriptions and examples are intended only to explain the present invention in detail and are not intended to limit the present invention. Any modifications and variations made to the present invention within the spirit of the present invention and the scope of protection of the claims shall be included within the scope of protection of the present invention.

[0049] Meanwhile, to facilitate understanding of the embodiments of the present invention, each operation is described as multiple discrete operations; however, the order of description does not represent the order in which the operations are performed.

[0050] In the description of the embodiments of the present invention, the sample measurement space is represented using three-dimensional coordinates (xyz) based on spatial directions, and letters such as I, H, and m are used to represent variables. This representation is intended solely to facilitate discussion and does not limit the application of the present invention. I, H, and m can take any value, such as 1, 2, or 3.

[0051] The method of the present invention is shown in Figure 1. An OCTA continuous imaging method 1: performing continuous OCT scanning imaging on the target tissue area to obtain OCT signals;

[0052] One approach to continuous OCTA imaging is to use one of the following methods:

[0053] Time domain OCT imaging method by changing the optical path length of the reference arm through scanning;

[0054] Alternatively, a spectral domain OCT imaging method using a spectrometer to record spectral interference signals;

[0055] Alternatively, a swept-frequency OCT imaging method is used to record spectral interference signals using a swept-frequency light source.

[0056] Method 2 for calculating the scattering coefficients of blood and surrounding tissue: generating a three-dimensional scattering coefficient and three-dimensional microvascular distribution of the target tissue region based on the OCT signal, and separately calculating the blood scattering coefficient (BOC) and the tissue scattering coefficient (TOC) based on the three-dimensional scattering coefficient and the three-dimensional microvascular distribution;

[0057] A method 2 for calculating the scattering coefficient of blood and surrounding tissues, specifically comprising:

[0058] The depth attenuation characteristics of OCT are used to extract the three-dimensional scattering coefficient of the target tissue area in the OCT signal21;

[0059] Utilizing the depth attenuation characteristics of OCT, the three-dimensional scattering coefficient of the target tissue area in the OCT signal is extracted21, specifically including:

[0060] First, the system noise of the OCT signal in the depth direction is removed to obtain the denoised OCT signal;

[0061] Then, the attenuation of the denoised OCT signal along the depth direction is compensated using a Gaussian fitting model to obtain a compensated OCT signal;

[0062] Finally, the scattering coefficient of the target tissue area is calculated based on the variation characteristics of the compensated OCT signal in the depth direction.

[0063] More specifically:

[0064] First, the system noise of the OCT signal in the depth direction is removed to obtain the denoised OCT signal. Then, the attenuation of the denoised OCT signal along the depth direction is compensated using a Gaussian fitting model to obtain the compensated OCT signal. Finally, the scattering coefficient of the target tissue area is calculated based on the variation characteristics of the compensated OCT signal in the depth direction.

[0065] The denoising operation in the above operation is achieved by the following formula: I(z)=I inital (z)-N(z);

[0066] Among them, I initalis the raw OCT signal intensity at depth z, N(z) is the average system noise at depth z, and I(z) is the denoised OCT signal intensity at depth z. The system average noise is extracted by first collecting the device noise by blocking the sample arm of the blood glucose test device. The system noise in each XY plane is then averaged along the depth direction to obtain a one-dimensional depth-wise distribution of the average system noise.

[0067] The compensation operation in the above operation is achieved through the following formula:

[0068] Among them, I correct is the compensated OCT signal intensity, S(z) is the Gaussian fitting model, which is obtained by Gaussian fitting the attenuation of OCT signal intensity along the depth direction. S(z)=exp(-z 2 / σ 2 );

[0069] Here, σ is the exponential attenuation parameter of the OCT signal intensity in the depth direction. The specific measurement method is: a sensitivity roll-off test is performed on the non-invasive blood glucose monitoring device, and the attenuation of the signal intensity per unit depth in the depth direction is calculated by Gaussian fitting, which is the attenuation parameter σ.

[0070] In the above operation, the three-dimensional scattering coefficient of the target area is calculated from the OCT depth attenuation signal using the following formula:

[0071] Where x represents the fast scan direction in 3D OCT scanning, y represents the slow scan direction in 3D OCT scanning, and z represents the depth direction in 3D OCT scanning, which is perpendicular to the plane formed by the fast and slow scan directions. μ(x,y,z) is the optical scattering coefficient at the pixel position (x,y,z), I(x,y,z) represents the compensated OCT signal intensity at the pixel position (x,y,z), n is the refractive index of the tissue area being measured, and Δz is the actual size of each pixel in air in the depth direction.

[0072] Generate three-dimensional microvascular distribution in the target tissue area based on OCT signals 22;

[0073] Generating a three-dimensional microvessel distribution 22 of the target tissue region according to the OCT signal specifically includes: analyzing the amplitude or phase or the amplitude and phase of the OCT signal using an OCT blood flow signal extraction method to obtain the three-dimensional microvessel distribution of the target tissue region.

[0074] OCT blood flow signal extraction methods include: differential calculation, speckle variance calculation, decorrelation calculation or feature decomposition calculation.

[0075] After the three-dimensional microvascular distribution image is binarized, the blood vessel mask and tissue mask are obtained. The blood scattering coefficient BOC and tissue scattering coefficient TOC 23 are obtained by multiplying the three-dimensional scattering coefficient with the blood vessel mask and tissue mask respectively.

[0076] After binarizing the three-dimensional microvascular distribution image, the blood vessel mask and tissue mask are obtained. The blood scattering coefficient BOC and tissue scattering coefficient TOC 23 are obtained by multiplying the three-dimensional scattering coefficient with the blood vessel mask and tissue mask, respectively. Specifically,

[0077] A suitable threshold is selected to perform a binarization operation on the three-dimensional microvascular distribution (i.e., the three-dimensional microvascular matrix). Specifically, the pixel values ​​of all blood flow areas in the three-dimensional microvascular distribution are set to 1, and the pixel values ​​of non-blood flow areas are set to 0. This is to obtain a blood vessel mask. The blood flow scattering coefficient is then extracted by performing a dot product operation on the blood vessel mask and the three-dimensional scattering coefficient.

[0078] The pixel values ​​of all blood flow areas in the three-dimensional microvascular distribution are set to 0, and the pixel values ​​of the non-blood flow areas are set to 1 to obtain the tissue mask. The tissue mask is then multiplied with the three-dimensional scattering coefficient to extract the tissue scattering coefficient.

[0079] A blood glucose concentration calibration and calculation method 3: Collect blood samples and use an electrochemical method to quickly measure the reference blood glucose concentration at two times, t1 and t2. It is recommended that t1 and t2 be before and after a meal, respectively. The sampling frequency can be increased to improve calibration accuracy. The reference blood glucose concentration and blood scattering coefficient at the same time are paired and then linearly fitted. The first fitting formula satisfies: BGC = a*BOC + b. The first and second fitting parameters a and b are obtained, thereby obtaining the reference blood glucose concentration at all times.

[0080] Specifically, blood glucose concentrations are measured using a commercial blood glucose meter or other means at two times, t1 and t2, before and after a meal, serving as reference blood glucose concentrations. The OCTA-measured blood scattering coefficients at these times are paired with the reference blood glucose concentrations. A linear relationship is established between the two using a linear fitting method: BGC = a*BOC + b, yielding fitting parameters a and b. This linear relationship can be used to map the OCTA-measured blood scattering coefficients to the actual blood glucose concentration.

[0081] A calibration and calculation method 4 for interstitial fluid glucose concentration: The peak time difference ΔT of the blood scattering coefficient (BOC) and the tissue scattering coefficient (TOC) is used. The reference blood glucose concentrations at time points t1 and t2 are paired with the corresponding tissue scattering coefficients at a delay time of ΔT, and then linear fitting is performed. The second fitting formula satisfies: IGC = c*TOC + d. The third and fourth fitting parameters c and d are obtained, thereby obtaining the interstitial fluid glucose concentration IGC at all times.

[0082] Specifically, the lag time ΔT of interstitial fluid glucose concentration relative to blood glucose concentration is calculated based on the peak times of the BOC and TOC postprandial curves. Reference blood glucose concentrations at times t1 and t2 are paired with the tissue scattering coefficients (TSCs) delayed by ΔT, respectively. A linear relationship is established using a linear fitting method: IGC = c*TOC + d, yielding fitting parameters c and d. This linear relationship, based on the tissue scattering coefficients measured by OCTA, can be used to map interstitial fluid glucose concentration.

[0083] In the embodiments disclosed in the present invention, calibration of an individual's blood glucose concentration and interstitial fluid glucose concentration is achieved through an oral glucose tolerance test.

[0084] Subjects underwent an oral glucose tolerance test. Reference blood glucose concentrations were measured rapidly using chemical methods at baseline (t1) and 20 minutes after ingestion of a standard glucose solution (t2). The reference blood glucose concentrations at the same time points were paired with the blood scattering coefficient (BGC). Linear fitting was used to obtain the fitting parameters a and b, and thus the blood glucose concentration (BGC) at all times.

[0085] Using an OCTA device, subjects undergoing an oral glucose tolerance test were continuously monitored for BOC and TOC, and the peak time difference ΔT was obtained. Reference blood glucose concentrations at baseline (t1) and 20 minutes after ingestion of a standard glucose solution (t2) were paired with the tissue scattering coefficients at the respective ΔT delays. Linear fitting was used to obtain the fitting parameters c and d, and thus the interstitial fluid glucose concentration (IGC) at all times was obtained.

[0086] Figure 2 shows a schematic diagram of the continuous, non-invasive optical coherence blood glucose detection device of the present invention. The main structure of the low-coherence interference measurement part of the device is an interferometer, which consists of a light source 11, a sample arm objective lens 17, an interference signal detection device 19 and a polarization controller 24. The light emitted by the light source 11 is divided into two light beams by a beam splitter 12: one beam enters the reference arm of the interferometer and is irradiated on the plane high-reflection mirror 14 through the reference arm collimator 13; the other beam enters the sample arm and is focused on the sample to be tested after being collimated 15 and reflected by the optical path; the sample 18 is placed at the focal plane of the sample arm objective lens 17. The light reflected back from the reference arm and the sample arm then interferes and is received by the interference signal detection device 19. For the fiber-optic optical path, a polarization controller 24 is used to adjust the polarization state of the light beam to maximize the signal interference effect.

[0087] Based on different methods of low-coherence interferometric detection signals, the continuous, non-invasive optical coherence blood glucose detection device shown in FIG2 specifically includes:

[0088] 1) Time Domain Measurement Device. Light source 11 uses broadband, low-coherence light. Plane mirror 14 is movable along the optical axis. Interference signal detection device 19 is a point detector. By moving plane mirror 14 to change the optical path length of the reference arm, the interference signal between the two arms is detected by point detector 19. Low-coherence interferometry detection of the scattered signal in the z-direction at a specific spatial depth allows for the acquisition of a sample volume in the depth spatial dimension.

[0089] 2) Spectral Domain Measurement Apparatus. Light source 11 uses broadband, low-coherence light, plane reflector 14 is stationary, and interference signal detection device 19 uses a spectrometer. The interference signal passes through a linear array camera in the spectrometer, which simultaneously records the interference spectrum. Fourier analysis is used to analyze the interference spectrum signal, and scattering information in the depth z direction is simultaneously acquired, thereby obtaining a sample volume in the depth dimension.

[0090] 3) Sweep-Frequency Measurement Device. Light source 11 uses a swept-frequency source, plane reflector 14 is fixed, and interference signal detection device 19 uses a point detector. The point detector records the low-coherence interference spectrum of the swept-frequency source in a time-sharing manner. The interference spectrum signal is sampled and analyzed using Fourier transforms, and scattering information in the depth z direction is acquired simultaneously, thereby obtaining a sample volume in the depth dimension.

[0091] The above-mentioned different measurement devices can be respectively combined with the OCTA optical coherence angiography method described in FIG1 to perform noise suppression and defect compensation, and enhance the real-time performance and spatial correspondence of the measurement.

[0092] Figure 3 illustrates an exemplary embodiment of the present invention disclosed herein. The continuous, non-invasive optical coherence blood glucose detection device includes a broadband swept-frequency light source 31, an optical fiber coupler 32 with an 80:20 splitting ratio, an optical fiber coupler 33 with a 50:50 splitting ratio, a first optical circulator 34, a second optical circulator 35, a polarization controller 36, a first collimator 40, a second collimator 37, a focusing lens 38, a scanning lens 42, a plane high-reflection mirror 39, a scanning galvanometer 41, a sample 43, a balanced detector 44, and a processor 45. Light emitted by the broadband swept-frequency light source 31 used in the present invention enters the optical fiber coupler 32 with an 80:20 splitting ratio through an optical fiber. The light emitted from the optical fiber coupler 32 is split into two sub-beams: one beam passes through the second optical circulator 35 and enters the second collimator 37. The parallel light emitted from the second collimator 37 is focused by the focusing lens 38 and irradiated onto the plane high-reflection mirror 39. Another beam of light passes through the first optical circulator 34 and enters the first collimator 40. It then passes through the scanning galvanometer 41 and scanning lens 42 before irradiating the sample 43. The light emitted from the second optical circulator 35 passes through the polarization controller 36 and, together with the light emitted from the first optical circulator 34, enters the 50:50 fiber coupler 33. The light emitted from the fiber coupler 33 enters the balanced detector 44 for interference signal acquisition and detection. The signal is then fed into the processor 45, which, combined with the light source's clock and trigger signals, performs further analysis and processing to produce an interference spectrum signal.

[0093] Figure 4 illustrates the blood glucose testing process according to an embodiment of the present invention. The test lasted 90 minutes, with the baseline period from -10 to 0 minutes, the glucose solution administered from 0 to 5 minutes, and the blood glucose recovery period from 5 minutes to 90 minutes. During the test, OCTA signal acquisition was performed every 5 minutes using the detection device described herein to obtain OCT signals from the target area for data processing. Blood glucose concentration was measured electrochemically every 10 minutes as a reference blood glucose level.

[0094] Figure 5 (A) shows a three-dimensional OCT structure diagram of the human fingernail area in the human finger skin imaging experiment, Figure 5 (B) shows a three-dimensional scattering coefficient distribution diagram of the human fingernail area, and Figure 5 (C) shows a three-dimensional OCTA blood flow angiography diagram of the human fingernail area.

[0095] Figure 6 shows a typical OCTA image from an embodiment of the present invention. Figure 6(A) shows a cross-sectional view of the OCT structure of the target region. The curve on the left shows the depth-wise attenuation characteristics of the OCT signal intensity in the region indicated by the white dashed line in Figure (A). Figure 6(B) shows a cross-sectional view of the scattering coefficient of the target region. Figures 6(C) and 6(D) show cross-sectional views of the OCTA angiography and OCTA mask of the target region.

[0096] FIG6(E) is a maximum projection image of OCTA blood flow angiography, wherein the white dotted lines indicate the positions of the cross-sectional images of FIG6(A) to FIG6(D).

[0097] Figure 7 is a graph showing the relationship between the optical blood glucose concentration, optical interstitial fluid sugar concentration and reference blood glucose concentration obtained by the method of the present invention when performing an oral glucose tolerance test in an embodiment of the present invention over time. Figure 7 (A) is a graph showing the change of the optical blood glucose concentration and the reference blood glucose concentration over time during the oral glucose tolerance test experiment. The changes in the optical blood glucose concentration and the reference blood glucose concentration are completely synchronized without any obvious delay time. Figure 7 (B) is a correlation characterization of the optical blood glucose concentration and the reference blood glucose concentration, with R = 0.92. Figure 7 (C) is a graph showing the change of the optical interstitial fluid sugar concentration and the reference blood glucose concentration over time during the oral glucose tolerance test experiment, where the optical interstitial fluid sugar concentration has a lag time of about 15 minutes relative to the reference blood glucose concentration. Figure 7 (D) is a correlation characterization of the optical interstitial fluid sugar concentration and the reference blood glucose concentration. Due to the influence of the lag time, the correlation R value between the two is 0.15, but after correcting for the lag time, the correlation R value increases to 0.89.

[0098] The above experimental results fully demonstrate that the method and device described in the present invention can achieve continuous and non-invasive optical blood glucose measurement, and the detection process is friendly and convenient, which has inspiring significance for non-invasive blood glucose monitoring.

Claims

1. A non-invasive continuous blood glucose measurement method based on OCTA, characterized in that: include: An OCTA continuous imaging method (1): performing continuous OCT scanning imaging on a target tissue region to obtain an OCT signal; A method for calculating the scattering coefficient of blood and surrounding tissue (2): generating a three-dimensional scattering coefficient and a three-dimensional microvascular distribution of a target tissue region according to an OCT signal, and obtaining a blood scattering coefficient and a tissue scattering coefficient based on the three-dimensional scattering coefficient and the three-dimensional microvascular distribution by separate calculation; A method for calibrating and calculating blood glucose concentration (3): collecting blood samples and measuring reference blood glucose concentrations at two times t1 and t2 using an electrochemical method, pairing the reference blood glucose concentrations and blood scattering coefficients at the same time and then performing linear fitting to obtain blood glucose concentrations BGC at all times; A method for calibrating and calculating interstitial fluid glucose concentration (4): pairing the reference blood glucose concentration at two times t1 and t2 with the tissue scattering coefficient at the corresponding delay time ΔT and then performing linear fitting to obtain the interstitial fluid glucose concentration IGC at all times.

2. The non-invasive continuous blood glucose measurement method based on OCTA according to claim 1, characterized in that: The OCTA continuous imaging method (1) is one of the following methods: Time domain OCT imaging method by scanning and changing the optical path length of the reference arm; or a spectral domain OCT imaging method using a spectrometer to record spectral interference signals; Alternatively, a swept-frequency OCT imaging method is used that uses a swept-frequency light source to record spectral interference signals.

3. The non-invasive continuous blood glucose measurement method based on OCTA according to claim 1, characterized in that: The method (2) for calculating the scattering coefficient of blood and surrounding tissues specifically comprises: Using the depth attenuation characteristics of OCT, the three-dimensional scattering coefficient of the target tissue area in the OCT signal is extracted (21); Generate three-dimensional microvascular distribution in the target tissue area based on OCT signals (22); After the three-dimensional microvascular distribution image is binarized, the vascular mask and tissue mask are obtained. The blood scattering coefficient and tissue scattering coefficient are obtained by multiplying the three-dimensional scattering coefficient with the vascular mask and tissue mask respectively (23).

4. The non-invasive continuous blood glucose measurement method based on OCTA according to claim 3, characterized in that: The method utilizes the depth attenuation characteristics of OCT to extract the three-dimensional scattering coefficient (21) of the target tissue area in the OCT signal, specifically comprising: First, the system noise of the OCT signal in the depth direction is removed to obtain the denoised OCT signal; Then, the attenuation of the denoised OCT signal along the depth direction is compensated using the Gaussian fitting model. Compensation, to obtain the compensated OCT signal; Finally, the scattering coefficient of the target tissue area is calculated based on the variation characteristics of the compensated OCT signal in the depth direction.

5. The non-invasive continuous blood glucose measurement method based on OCTA according to claim 3, characterized in that: The method of generating the three-dimensional microvessel distribution (22) of the target tissue region based on the OCT signal specifically includes: analyzing the amplitude or phase or the amplitude and phase of the OCT signal using the OCT blood flow signal extraction method to obtain the three-dimensional microvessel distribution of the target tissue region.

6. The non-invasive continuous blood glucose measurement method based on OCTA according to claim 5 is characterized in that: The OCT blood flow signal extraction method includes: differential calculation, speckle variance calculation, decorrelation calculation or feature decomposition calculation.

7. The non-invasive continuous blood glucose measurement method based on OCTA according to claim 3, characterized in that: After the three-dimensional microvascular distribution is binarized, a blood vessel mask and a tissue mask are obtained, and after the three-dimensional scattering coefficient is dot-multiplied with the blood vessel mask and the tissue mask, a blood scattering coefficient and a tissue scattering coefficient (23) are obtained, which specifically includes: After setting the pixel values ​​of all blood flow areas in the three-dimensional microvascular distribution to 1 and the pixel values ​​of the non-blood flow areas to 0, a blood vessel mask is obtained, and then the blood flow scattering coefficient is extracted after performing a dot multiplication operation on the blood vessel mask and the three-dimensional scattering coefficient; The pixel values ​​of all blood flow areas in the three-dimensional microvascular distribution are set to 0, and the pixel values ​​of non-blood flow areas are set to 1 to obtain the tissue mask. The tissue mask is then multiplied with the three-dimensional scattering coefficient to extract the tissue scattering coefficient.

8. A non-invasive continuous blood glucose measurement device based on OCTA for implementing any one of claims 1 to 7, characterized in that: The device includes: an optical coherence tomography device for continuous OCT imaging of a target area; One or more processors for dynamic real-time processing of data.

9. The non-invasive continuous blood glucose measurement device based on OCTA according to claim 8, characterized in that: The optical coherence tomography device is one of the following: Includes low coherence broadband light source and interferometer, point detector; or including a low-coherence broadband light source and an interferometer or spectrometer; or including a swept broadband light source and an interferometer, a point detector; Or other devices capable of achieving optical coherence tomography.

Images