Multifunctional sweep frequency OCT imaging system and method

Abstract

This invention discloses a multifunctional swept-frequency OCT imaging system and method. The system includes a swept-frequency light source, a 1*2 broadband fiber coupler, a reference arm module, a polarization modulation module, a three-channel broadband fiber circulator, a probe scanning module, a polarization-sensitive detection module, a high-speed data acquisition card, and a data processing module. The method includes a weighted angiography signal generation method, a threshold angiography signal generation method, and a high-precision birefringence (or phase delay) calculation method. This invention can achieve high-quality imaging of structures, blood vessels, and polarization; shorten the data acquisition time for vascular imaging; improve the contrast and signal-to-noise ratio of angiography images; increase data processing speed; improve the contrast between noise and blood vessels, avoiding the possibility of false blood vessels being misjudged; and improve the measurement accuracy of birefringence (or phase delay).

Description

Technical Field

[0001] This invention belongs to the field of optical imaging technology, specifically relating to a multifunctional swept-frequency OCT imaging system and method. Background Technology

[0002] Optical coherence tomography (OCT) utilizes the basic principle of low-coherence optical interference to detect backscattered signals at different depths of a sample. Through scanning, two-dimensional or three-dimensional images of the sample can be obtained, offering advantages such as low-destructiveness, high resolution, non-invasiveness, and real-time imaging. To improve the specificity and sensitivity of the detected objects, simple structural imaging OCT is gradually evolving towards OCT that integrates functional and structural imaging, including polarization-sensitive OCT (PS-OCT) and OCT angiography (OCTA).

[0003] OCTA imaging technology distinguishes between static biological tissues and vascular regions by calculating the changes between continuously acquired OCT signals. Based on the components of the calculated signals, OCTA technology can be divided into intensity-based, phase-based, and complex-signal-based angiography. Among them, complex-signal-based OCTA technology can improve blood flow detection sensitivity and motion contrast, and has gained increasing popularity among researchers. However, it is susceptible to phase artifacts introduced by overall motion and system instability, especially in swept-frequency OCT systems. In 2016, GANGJUN LIU proposed a phase difference gradient method (PGA) for angiography imaging, which not only effectively removes phase errors introduced by depth-independent displacement but also reduces artifacts introduced by swept-frequency light source instability. By combining the PGA method with the spectral amplitude decorrelation method (SSADA), called the spectral amplitude and phase gradient method (SSAPGA), imaging performance and quality are further improved. However, this method has the following problems:

[0004] (1) Although the spectral method can improve the signal-to-noise ratio of the image, it greatly increases the data processing time.

[0005] (2) The spectral method reduces axial resolution, which in turn reduces the sensitivity of blood flow detection and the motion contrast of the image.

[0006] (3) The SSAPGA method cannot eliminate phase artifacts introduced by depth-related axial subpixel displacement.

[0007] Furthermore, the signal-to-noise ratio (SNR) of OCTA signals is closely related to that of OCT signals. To eliminate spurious vessels introduced by low SNR signals, pixel-by-pixel weighting of the angiography image is typically performed using the corresponding OCT scattering intensity map, or intensity thresholds are established to remove artifacts. However, traditional methods have the following problems:

[0008] (1) Although the pixel-by-pixel weighting method preserves all signals for pathologists to analyze, it does not significantly improve the contrast between blood vessels and noise, and the possibility of false blood vessels being misjudged is still very high.

[0009] (2) The process of determining the intensity threshold is complex. On the one hand, the background signal without a sample can be acquired before or after data acquisition to estimate the noise intensity, but this method reduces the efficiency of data acquisition. Alternatively, the acquired OCT image can be segmented to obtain pure noise regions to estimate the noise intensity, but this method is not suitable for samples with constantly changing surface positions.

[0010] (3) The intensity threshold is often chosen based on experience, which lacks rationality. It cannot guarantee that the true OCTA signal is preserved while removing noise.

[0011] As another functional extension of OCT, PS-OCT can be combined with OCTA to improve vascular connectivity and density by eliminating polarization artifacts. More importantly, it can extract polarization characteristics related to the microscopic and even ultramicroscopic structure of the sample. For example, birefringence (or phase delay) calculated by decomposing the eigenvalues ​​of the Jones matrix can be used to detect the distribution and changes of structure and composition, showing unique advantages, especially in the biomedical field. However, the measurement accuracy of traditional methods for solving birefringence (or phase delay) is highly dependent on the signal-to-noise ratio of the OCT signal, especially in the low signal-to-noise ratio region, where the measured values ​​deviate significantly from the true values.

[0012] Sensitivity is a crucial factor in both structural and functional OCT systems, reaching its maximum only when the system reaches the shot noise limit. Since the sample arm optical power is significantly lower than the reference arm optical power, the system noise is primarily determined by the average optical power of the reference arm. Theoretically, gradually increasing the reference optical power until the system reaches the shot noise limit can achieve optimal sensitivity. However, due to manufacturing errors in the optical components, the power entering the two input ports of the balanced detector is not equal, resulting in a DC term remaining at the differential output port of the balanced detector. This DC term increases with increasing reference optical power, inevitably exceeding the voltage range detectable by the data acquisition card, thus preventing the system from achieving optimal sensitivity and consequently reducing the quality of angiography and polarization imaging.

[0013] The aforementioned low performance issues of the system and imaging methods severely limit the commercialization and clinical application of multifunctional swept-frequency OCT imaging systems. Summary of the Invention

[0014] In view of this, the present invention provides a multifunctional swept-frequency OCT imaging system and method, which can improve system sensitivity and achieve high-quality imaging of structures, blood vessels, and polarization; shorten the data acquisition time for vascular imaging; improve the contrast and signal-to-noise ratio of angiography images; improve data processing speed; improve the contrast between noise and blood vessels, avoiding the possibility of false blood vessels being misjudged; and improve the measurement accuracy of birefringence (or phase delay).

[0015] The technical solution for implementing the present invention is as follows:

[0016] A multifunctional swept-frequency OCT imaging system includes a swept-frequency light source, a 1*2 broadband fiber coupler, a reference arm module, a polarization modulation module, a three-channel broadband fiber circulator, a probe scanning module, a polarization-sensitive detection module, a high-speed data acquisition card, and a data processing module.

[0017] The system employs a half-fiber structure to enhance stability and noise immunity. A swept-frequency light source outputs low-coherence light with a sweep rate of M, a center wavelength of λ0, and a bandwidth of Δλ. This light is split into sample light and reference light by a 1*2 broadband fiber coupler at a power ratio A / B, and then transmitted to the polarization modulation module and reference arm module, respectively. The sample light is modulated by the polarization modulation module to generate polarized light with defined properties, and then transmitted to the probe scanning module via a three-channel broadband fiber circulator and focused onto the sample. The backscattered light returning from the sample is collected by the probe scanning module and transmitted to the polarization-sensitive detection module via the three-channel broadband fiber circulator. The reference light undergoes optical path modulation by the reference arm module, is transmitted via single-mode fiber to the polarization-sensitive detection module, and interferes with the sample backscattered light. Finally, a high-speed data acquisition card acquires the interference signal and uploads it to the data processing module for subsequent data processing.

[0018] Furthermore, the polarization-sensitive detection module includes a 2*2 broadband fiber coupler, an aperture, two polarization-sensitive beam splitters, and two balanced detectors. The two input ports of the 2*2 broadband fiber coupler are connected to the reference arm module via single-mode fiber and to the probe scanning module via a three-channel broadband fiber circulator, respectively, so that the reference light and the sample backscattered light interfere at the 2*2 broadband fiber coupler. The output port with higher output power of the 2*2 broadband fiber coupler outputs the interference signal to the polarization-sensitive beam splitter via the aperture, while the output port with lower output power directly outputs the interference signal to the polarization-sensitive beam splitter. The two polarization-sensitive beam splitters orthogonally separate the interference signals, and the interference spectra with the same polarization state are received by the same balanced detector.

[0019] Furthermore, the actual splitting ratio of the 2*2 broadband fiber coupler is C / D (ideally, it can be split at a power ratio of 1:1). Due to manufacturing errors, C≠D.

[0020] A multifunctional swept-frequency OCT imaging method includes data preprocessing, Jones synthesis, angiography generation, and angiography weighting. Data preprocessing refers to the routine processing of R (R≥2) interference spectra continuously acquired at the same lateral position of the sample at time intervals Δt, including DC term removal, dispersion compensation, zero-padding, Fourier transform, and spatial alignment using sub-pixel cross-correlation, thereby obtaining a Jones matrix. Jones synthesis refers to the coherent synthesis of the four obtained Jones components to obtain a sensitivity-enhanced OCT signal.

[0021] Angiography generation involves calculating the gradient of the phase difference and decorrelation of the amplitude of the obtained complex signal OCT, and finally obtaining the difference value of the newly formed complex signal along the depth direction within a Gaussian window (window size is 2L+1, where L is any positive integer), thereby generating a high motion contrast angiography signal.

[0022] Angiography weighting refers to the weighting of obtained angiography signals using an automatic weighting method based on DOPU signals.

[0023] Further, the weighted angiography processing specifically involves: first, converting the obtained Jones vector (the column vector of the Jones matrix obtained from data preprocessing) into a Stokes vector, and using the Stokes vector to calculate the DOPU signal at an initial spatial size k*l (where k and l are both integers greater than 1); then obtaining the histogram distribution of the pixel values ​​in the DOPU image and the corresponding probability density curve; when the DOPU value where the minimum value of the total probability density curve is located is less than the confidence interval α (0 < α < 1) under the noise probability density curve, it is considered that the preset threshold condition has been met at that spatial size; otherwise, the spatial size is gradually increased until the threshold condition is met, and the DOPU signal at that spatial size is calculated; finally, the DOPU image that meets the threshold condition is weighted pixel by pixel with the angiography image to obtain a weighted angiography image with high vessel / noise contrast.

[0024] A multifunctional swept-frequency OCT imaging method includes data preprocessing, Jones synthesis, angiography generation, and angiography thresholding. Data preprocessing refers to the routine processing of R (R≥2) interference spectra continuously acquired at the same lateral position of the sample at time intervals Δt, including DC term removal, dispersion compensation, zero-padding, Fourier transform, and spatial alignment using subpixel cross-correlation, thereby obtaining a Jones matrix. Jones synthesis refers to the coherent synthesis of the four obtained Jones components to obtain a sensitivity-enhanced OCT signal.

[0025] Angiography generation involves calculating the gradient of the phase difference and decorrelation of the amplitude of the obtained complex signal OCT, and finally obtaining the difference value of the newly formed complex signal along the depth direction within a Gaussian window (window size 2L+1), thereby generating a high motion contrast angiography signal.

[0026] Angiography thresholding refers to the thresholding of obtained angiography signals using an automatic thresholding method based on DOPU signals.

[0027] Further, the angiography thresholding process specifically involves: first, converting the obtained Jones vector into a Stokes vector, and then using the Stokes vector to calculate the DOPU signal at the initial spatial size k*l; then, obtaining the histogram distribution of the pixel values ​​in the DOPU image and the corresponding probability density curve; when the DOPU value where the minimum value of the total probability density curve is located is less than the confidence interval α under the noise probability density curve, it is considered that the preset threshold condition has been met at that spatial size; otherwise, the spatial size is gradually increased until the threshold condition is met, and the DOPU value where the minimum value of the total probability density curve is located is used as the DOPU threshold; pixels exceeding the DOPU threshold retain the angiography signal, while pixels below the DOPU threshold have their angiography signal set to 0, thereby filtering out pseudo-vascular signals introduced by low signal-to-noise ratio.

[0028] A multifunctional swept-frequency OCT imaging method includes data preprocessing, Jones matrix spatial averaging, Jones matrix calibration, matrix transformation, Mueller matrix temporal averaging, and polar decomposition. Data preprocessing involves performing conventional processing on K (K≥2) interferometric spectra continuously acquired from the same lateral position of the sample, including DC term removal, dispersion compensation, zero-padding, Fourier transform, and spatial alignment using subpixel cross-correlation, to obtain the Jones matrix. An adaptive averaging method is used to spatially average the obtained Jones matrix to improve the signal-to-noise ratio. Then, a calibrated Jones matrix is ​​obtained using the sample surface as a reference to eliminate the influence of birefringence on the phase delay measurement results. Matrix transformation involves converting the obtained K calibrated Jones matrices into K Mueller matrices.

[0029] The time averaging of the Mueller matrix refers to summing and averaging the K transformed Mueller matrices to obtain a Mueller matrix containing debiasing information caused by system instability and speckle noise. Finally, the Mueller matrix is ​​decomposed to remove the debiasing information introduced by system instability and speckle noise, thereby obtaining a high-precision phase delay value.

[0030] Beneficial effects:

[0031] 1. This invention provides a high-sensitivity, multi-functional swept-frequency OCT imaging system. First, an aperture is added to the port with higher output power of a 2*2 fiber coupler with an actual splitting ratio of C / D (C≠D). Then, using a plane mirror as the test sample, the average power of the reference light is gradually increased while the aperture diameter is decreased to prevent the differential output voltage of the balance detector from exceeding the detectable voltage range of the data acquisition card. The sensitivity at the plane mirror and the average optical power of the reference light are recorded until the optimal sensitivity of the system is found. This invention avoids oversaturation of the data acquisition card and achieves the system's highest sensitivity, improving the imaging quality of sample structure, polarization, and angiography.

[0032] 2. This invention provides an angiography imaging method that improves immunity to phase artifacts introduced by overall motion and system instability, while also enhancing motion contrast and data processing speed. This method combines phase difference gradient calculation and intensity decorrelation methods, and calculates the difference values ​​within a Gaussian window (size 2L+1) along the depth direction to obtain the angiography signal. This method preserves full-spectrum information without sacrificing sensitivity to blood flow detection, thus improving motion contrast. Furthermore, the summation along the depth direction immunizes against phase artifacts introduced by depth-related subpixel displacement. Compared to the spectral splitting method, this method significantly improves data processing speed.

[0033] 3. This invention provides an adaptive weighting method to improve the contrast between noise and blood vessels. First, the measured polarization information is converted into Stokes vectors. Then, DOPU maps at different spatial dimensions are calculated using the Stokes vectors, and the histogram of each frame of the DOPU map corresponding to OCTA is analyzed until a threshold condition that can distinguish between signal and noise is reached. Finally, the DOPU map at this spatial dimension is calculated and weighted pixel-by-pixel to the angiography image. Compared with traditional methods, this method can improve the contrast between noise and blood vessels and enhance the recognition of pseudo-blood vessels introduced by low signal-to-noise ratio signals.

[0034] 4. This invention provides an adaptive threshold masking method for angiography images. First, the measured polarization information is converted into Stokes vectors. Then, DOPU maps at different spatial dimensions are calculated using the Stokes vectors, and the histogram of each frame's DOPU map corresponding to OCTA is analyzed until a threshold condition that can distinguish between signal and noise is reached, and the corresponding DOPU value is obtained. Finally, pixels with DOPU values ​​greater than this threshold retain the OCTA signal, while pixels with DOPU values ​​less than this threshold have an OCTA signal of 0. Compared with traditional methods, this method can eliminate noise to the maximum extent without losing the OCTA signal. More importantly, this method avoids the complex process of noise intensity estimation.

[0035] 5. This invention provides a high-precision method for calculating birefringence (or phase delay). First, multiple calibrated Jones matrices acquired from the same location are converted into Mueller matrices. Then, the multiple Mueller matrices are summed and averaged to obtain a Mueller matrix containing debiasing information. Finally, the debiasing matrix introduced by noise is removed using the polar decomposition method, thereby obtaining a high-precision birefringence (or phase delay) value. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of an embodiment of the multifunctional swept-frequency OCT imaging system of the present invention.

[0037] Among them: 1-Sweep laser, 2-1*2 broadband fiber coupler, 3-Reference arm module, 4-Polarization modulation module, 5-Three-channel broadband fiber circulator, 6-Probe scanning module, 7-Polarization sensitive detection module, 71-2*2 broadband fiber coupler, 72-Fiber collimator, 73-Polarization sensitive beam splitter prism, 74-Aperture, 75-Balance detector, 8-High-speed data acquisition card, 9-Data processing and control module.

[0038] Figure 2 The curve shows the system sensitivity as a function of the average power of the reference light.

[0039] Figure 3 The diagram shows the flowchart for generating weighted angiography signals.

[0040] Figure 4 The image shown is a diagram of the experimental results of fluid-weighted angiography.

[0041] Figure 5 The diagram shows the flowchart for generating threshold angiography signals.

[0042] Figure 6 The figure shown is an experimental result of threshold angiography of fluid.

[0043] Figure 7 The flowchart shown is for high-precision birefringence (or phase delay) solution.

[0044] Figure 8 The image shown is a result of the phase delay of the nail. Detailed Implementation

[0045] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0046] like Figure 1The diagram shows an embodiment of a high-sensitivity multi-functional swept-frequency OCT imaging system, including a swept-frequency light source 1, a 1*2 10 / 90 broadband fiber coupler 2, a reference arm module 3, a polarization modulation module 4, a three-channel broadband fiber optic circulator 5, a probe scanning module 6, a polarization-sensitive detection module 7, a high-speed data acquisition card 8, and a data processing and system control module 9.

[0047] The swept frequency light source can generate low-coherence light with a swept frequency rate of 200kHz, a center wavelength of 1310nm, and a bandwidth of 100nm.

[0048] The polarization-sensitive detection module includes: a 2*2 broadband fiber coupler 71, an aperture 74, a polarization-sensitive beam splitter 731, a polarization-sensitive beam splitter 732, a balanced detector 751, a balanced detector 752, and fiber collimating lenses 72 (721-726). The actual splitting ratio of the 2*2 broadband fiber coupler is 47.5 / 52.5.

[0049] The high-speed data acquisition card uses the ATS9371 and can measure voltages within a range of ±400mV.

[0050] Using a plane mirror as the test sample, the average power of the reference light was gradually increased while the aperture of the aperture was decreased, so that the interference signal approached but did not exceed the detection saturation value. The average power value of the reference light and the corresponding system sensitivity value were recorded each time, thus obtaining a curve showing the relationship between sensitivity and average power of the reference light. Figure 2 As shown.

[0051] Figure 3 This diagram illustrates the flowchart for generating a weighted angiography signal. It includes data preprocessing, Jones synthesis, angiography generation, and weighted angiography processing. Data preprocessing involves performing routine processing on R (R≥2) interference spectra acquired continuously at the same lateral position of the sample at time intervals Δt. This includes removing DC terms, dispersion compensation, zero-padding, Fourier transform, and spatial alignment using subpixel cross-correlation to obtain the Jones matrix. The four Jones components are then coherently synthesized using the following formula to obtain a sensitivity-enhanced OCT signal.

[0052]

[0053] in,

[0054]

[0055]

[0056]

[0057] The gradient of the phase difference and the decorrelation of the amplitude of the obtained sensitive enhanced complex signal OCT are calculated. Finally, the difference value of the newly formed complex signal along the depth direction is obtained within a Gaussian window (the window size is 2L+1, where L is any positive integer), thereby generating a high motion contrast angiography signal.

[0058] Angiography weighting refers to the weighting of acquired angiography signals using an automatic weighting method based on DOPU signals. First, the obtained Jones vector is converted into a Stokes vector using the following formula:

[0059]

[0060] Then, the DOPU signal under the initial spatial size k*l is calculated.

[0061]

[0062] The histogram distribution of pixel values ​​in the DOPU image and the corresponding probability density curve are obtained. When the DOPU value where the minimum value of the total probability density curve is located is less than 97.5% of the confidence interval under the noise probability density curve, it is considered that the preset threshold condition has been met within that spatial size. Otherwise, the spatial size is gradually increased until the threshold condition is met, and the DOPU signal within that spatial size is calculated. Finally, the DOPU image that meets the threshold condition is weighted pixel-by-pixel with the angiography image to obtain a weighted angiography image with high vessel / noise contrast. Figure 4 The image shown is a weighted angiography obtained through a fluid experiment.

[0063] Figure 5 This diagram illustrates the flowchart of the angiography signal after threshold generation. It includes data preprocessing, Jones synthesis, angiography generation, and angiography threshold processing. Data preprocessing refers to the routine processing of R (R≥2) interference spectra continuously acquired at the same lateral position of the sample at time intervals Δt. This includes removing DC terms, dispersion compensation, zero-padding, Fourier transform, and spatial alignment using subpixel cross-correlation to obtain the Jones matrix. The four Jones components are then coherently synthesized using the following formula to obtain a sensitivity-enhanced OCT signal.

[0064]

[0065] in,

[0066]

[0067]

[0068]

[0069] The gradient of the phase difference and the decorrelation of the amplitude of the obtained sensitive enhanced complex signal OCT are calculated. Finally, the difference value of the newly formed complex signal along the depth direction is obtained within a Gaussian window (window size 2L+1), thereby generating a high motion contrast angiography signal.

[0070] Angiography thresholding refers to the thresholding of the obtained angiography signal using an automatic thresholding method based on DOPU signals. First, the obtained Jones vector is converted into a Stokes vector using the following formula:

[0071]

[0072] Then, the DOPU signal under the initial spatial size k*l is calculated.

[0073]

[0074] The histogram distribution of pixel values ​​in the DOPU image and the corresponding probability density curve are obtained. When the DOPU value where the minimum value of the total probability density curve is located is less than 97.5% of the confidence interval under the noise probability density curve, it is considered that the preset threshold condition has been met within that spatial size. Otherwise, the spatial size is gradually increased until the threshold condition is met, and the DOPU value where the minimum value of the total probability density curve is located is used as the DOPU threshold. Pixels exceeding the DOPU threshold retain the angiography signal, while pixels below the DOPU threshold have their angiography signal set to 0, thereby filtering out pseudo-vascular signals introduced by low signal-to-noise ratio. Figure 6 The image shown is an angiography image after the threshold was obtained through fluid experiments.

[0075] like Figure 7 The diagram shows the high-precision birefringence (or phase delay) solution flowchart. It includes data preprocessing, spatial averaging of the Jones matrix, Jones matrix calibration, matrix transformation, time averaging of the Mueller matrix, and pole decomposition. Data preprocessing involves performing conventional processing on K (K≥2) consecutively acquired interferometric spectra of the sample at the same transverse position. This includes removing the DC term, dispersion compensation, zero-padding, Fourier transform, and spatial alignment using subpixel cross-correlation to obtain the Jones matrix. An adaptive averaging method is used to spatially average the obtained Jones matrix to improve the signal-to-noise ratio.

[0076] Then, using the sample surface as a reference plane, the influence of birefringence within the system on the measurement of the accumulated phase delay value of the sample is eliminated, thereby obtaining the calibrated Jones matrix:

[0077]

[0078] The calculated K calibrated Jones matrices are converted into K Mueller matrices using the following formula.

[0079]

[0080] in,

[0081]

[0082] Then, the K transformed Mueller matrices are summed and averaged to obtain a Mueller matrix containing debiasing information caused by system instability and speckle noise. Finally, the Mueller matrix is ​​subjected to polar decomposition to remove the debiasing information introduced by system instability and speckle noise, thereby obtaining a high-precision phase delay value. Figure 8 The phase delay diagram shown is obtained using a fingernail as the test sample.

[0083] In summary, the above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A multifunctional swept-frequency OCT imaging system, characterized in that, It includes a swept frequency light source, a 1×2 broadband fiber coupler, a reference arm module, a polarization modulation module, a three-channel broadband fiber circulator, a probe scanning module, a polarization-sensitive detection module, a high-speed data acquisition card, and a data processing module; The low-coherence light output from the swept frequency source is split into sample light and reference light by a 1×2 broadband fiber coupler according to a power ratio A / B, and then transmitted to the polarization modulation module and the reference arm module, respectively. The sample light is modulated by the polarization modulation module to generate polarized light with set properties, and then transmitted to the probe scanning module by a three-channel broadband fiber circulator and focused onto the sample under test. The backscattered light returned from the sample is collected by the probe scanning module and transmitted to the polarization-sensitive detection module by the three-channel broadband fiber circulator. The reference light is optically path modulated by the reference arm module, and then transmitted to the polarization-sensitive detection module by a single-mode fiber and interferes with the sample backscattered light. Finally, the high-speed data acquisition card collects the interference signal and uploads it to the data processing module for subsequent data processing. The polarization-sensitive detection module includes a 2×2 broadband fiber coupler, an aperture, two polarization-sensitive beam splitters, and two balanced detectors. The two input ports of the 2×2 broadband fiber coupler are connected to the reference arm module via single-mode fiber and to the probe scanning module via a three-channel broadband fiber circulator, respectively, allowing the reference light and sample backscattered light to interfere at the 2×2 broadband fiber coupler. The output port with higher output power of the 2×2 broadband fiber coupler outputs the interference signal through the aperture to the polarization-sensitive beam splitter, while the output port with lower output power directly outputs the interference signal to the polarization-sensitive beam splitter. The two polarization-sensitive beam splitters orthogonally separate the interference signals, and interference spectra with the same polarization state are received by the same balanced detector. During the transmission of the interference signal, the average power of the reference light is gradually increased while the aperture is decreased to bring the interference signal close to but not exceeding the detection saturation value. The average power value of the reference light and the corresponding system sensitivity value are recorded each time, thus obtaining a curve showing the relationship between sensitivity and the average power of the reference light, until the optimal sensitivity of the system is found.

2. The multifunctional swept-frequency OCT imaging system as described in claim 1, characterized in that, The actual splitting ratio of the 2×2 broadband fiber coupler is: Due to process errors, .

3. A multifunctional swept-frequency OCT imaging method, characterized in that, This includes data preprocessing, Jones synthesis, angiography generation, and angiography weighting; among which, data preprocessing refers to continuously acquiring data from the same lateral position of the sample at time intervals. The R interference spectra are processed using conventional methods. This includes removing the DC term, dispersive compensation, zero-padding, Fourier transform, and spatial alignment using subpixel cross-correlation, thereby obtaining the Jones matrix; Jones synthesis refers to the coherent synthesis of the four obtained Jones components to obtain a sensitive OCT signal; Angiography generation involves calculating the gradient of the phase difference and decorrelating the amplitude of the acquired complex OCT signal. Finally, the difference value of the newly formed complex signal along the depth direction is obtained within a Gaussian window with a window size of [value missing]. This generates high motion contrast angiography signals; Angiography weighting refers to the weighting of obtained angiography signals using an automatic weighting method based on DOPU signals. The angiography weighting process specifically involves: first, converting the obtained Jones vector into a Stokes vector, and then using the Stokes vector to calculate the initial spatial dimensions. The DOPU signal is then analyzed; the histogram distribution of the pixel values ​​in the DOPU image and the corresponding probability density curve are obtained. The DOPU value at which the minimum value of the total probability density curve is located is less than the confidence interval under the noise probability density curve. If the preset threshold condition has been met within the specified spatial size, then the spatial size is gradually increased until the threshold condition is met, and the DOPU signal within that spatial size is calculated. Finally, the DOPU image that meets the threshold condition is weighted pixel by pixel with the angiography image to obtain a weighted angiography image with high vessel / noise contrast.

4. A multifunctional swept-frequency OCT imaging method, characterized in that, This includes data preprocessing, Jones synthesis, angiography generation, and angiography thresholding; among which, data preprocessing refers to continuously acquiring data from the same lateral position of the sample at time intervals. The R interference spectra are processed using conventional methods. This includes removing the DC term, dispersive compensation, zero-padding, Fourier transform, and spatial alignment using subpixel cross-correlation, thereby obtaining the Jones matrix; Jones synthesis refers to the coherent synthesis of the four obtained Jones components to obtain a sensitive OCT signal; Angiography generation involves calculating the gradient of the phase difference and decorrelating the amplitude of the acquired complex OCT signal. Finally, the difference value of the newly formed complex signal along the depth direction is obtained within a Gaussian window with a window size of [value missing]. This generates high motion contrast angiography signals; Angiography threshold processing refers to the threshold processing of obtained angiography signals using an automatic thresholding method based on DOPU signals. The angiography thresholding process specifically involves: first, converting the obtained Jones vector into a Stokes vector, and then using the Stokes vector to calculate the initial spatial dimensions. The DOPU signal is obtained; then the histogram distribution of the pixel values ​​in the DOPU image and the corresponding probability density curve are calculated. When the minimum value of the total probability density curve is located at a DOPU value that is less than the confidence interval under the noise probability density curve, the minimum value of the DOPU signal is determined. If the preset threshold condition has been met within the specified spatial size, then the spatial size is gradually increased until the threshold condition is met. The DOPU value at which the minimum total probability density curve is located is used as the DOPU threshold. Pixels exceeding the DOPU threshold retain the angiography signal, while pixels below the DOPU threshold have their angiography signal set to 0, thereby filtering out pseudo-vascular signals introduced by low signal-to-noise ratio.

Images