A method for intravascular polarization sensitive optical coherence tomography medium depolarization measurement

Abstract

This invention relates to a method for measuring the depolarization of a medium in intravascular polarization-sensitive optical coherence tomography (ICP-CT), comprising: setting the polarization state of the input light of a catheter polarization-sensitive optical coherence tomography system; setting the reference light in the H and V channels of the input light and the reference light in the system to have equal intensity; acquiring the electrical signals measured at the polarization diversity point in the form of a Jones matrix to obtain the average measurement Mueller matrix; calculating the standard deviation matrix of the reference region; obtaining the depolarization Mueller matrix of the reference region; extracting the block matrix of the average measurement Mueller matrix of the reference region; calculating the average double attenuation block matrix of the reference region; constructing the average double attenuation matrix of the reference region; obtaining the pseudo-average birefringence phase delay matrix of the reference region; extracting the average depolarization matrix of the reference region from the pseudo-average birefringence phase delay matrix of the reference region using matrix decomposition; obtaining the depolarization Mueller matrix of the target region; and calculating the medium depolarization coefficient.

Description

Technical Field

[0001] This invention relates to a catheter optical coherence tomography method. In particular, it relates to a method for depolarization measurement of intravascular polarization-sensitive optical coherence tomography media. Background Technology

[0002] Catheter-guided OCT is a high-resolution cross-sectional tomographic imaging technique. Catheter-guided PS-OCT, utilizing polarization imaging technology, can extract and record deep polarization information with high-resolution imaging, distinguishing different biological tissues. This technology can thus solve the medical challenge of identifying intravascular plaque lesions and rapidly assessing the stability of atherosclerotic plaques in vivo, in real-time. While existing OCT systems have reached a level where they can qualitatively determine the nature of tissue plaques, quantitative aspects, such as image clarity, imaging depth, and accuracy in determining tissue type, remain insufficient. Therefore, adopting and improving PS-OCT technology is a key direction for the development of OCT systems.

[0003] In catheter-based OCT, which uses light scattering intensity as its imaging mechanism, some tissues with similar scattering properties cannot be distinguished. However, when PS-OCT technology is introduced, these tissues become distinguishable due to differences in polarization properties. For example, related studies have shown differences in fibrous intermediate collagen or laminar arterial smooth muscle cells exhibiting birefringence effects; and differences in tissue fat content causing depolarization characteristics. Therefore, developing a catheter-based Polarization-sensitive OCT (PS-OCT) system capable of detecting polarization characteristics will further improve the accuracy of diagnosing the nature and structure of atherosclerotic plaques and correctly guide revascularization. Ding Zhenyang et al. from Tianjin University proposed using the Mueller matrix incoherent averaging method (CN202111079438.X), which can effectively demodulate the depolarization information of biological tissues. However, this method lacks the ability to distinguish noise-induced depolarization, and it is difficult to effectively solve problems such as blurred PS-OCT depolarization images caused by increased depolarization due to noise, and non-tissue plaques obscuring the actual plaque image. To address noise interference, S. Makita et al. proposed a DOPU calculation method that uses dual-channel mutual compensation to achieve noise correction, in order to more accurately characterize depolarization. Although this method achieves high noise immunity, the accuracy of this method was not systematically evaluated in this study [1]. In addition, this method has a high dependence on the input polarization state of the PS-OCT system [2]. N. Lippok et al. proposed a depolarization calculation method with lower polarization state dependence by combining a dual-polarization state PS-OCT system and the Mueller matrix, and verified this method in biological tissue imaging experiments [3,4]. This study showed the demodulation effect of the spatial averaging operator on depolarization, but ignored the noise analysis. Yamanari et al. used a dual-polarization state system and combined the quasi-maximum likelihood estimation method in the entropy calculation to get rid of the influence of noise on depolarization [5]. However, the accuracy of this method is highly dependent on the number of A-scans and requires the construction of a complex hardware system [2].

[0004] Related literature

[0005] [1] S. Makita, Y.-J. Hong, M. Miura, et al., "Degree of polarization uniformity with high noise immunity using polarization-sensitive optical coherence tomography," Opt. Lett. 39(24), 6783-6786 (2014).

[0006] [2]M.Yamanari,M.Mase,R.Obata,et al.,“Melanin concentration anddepolarization metrics measurement by polarization-sensitive opticalcoherence tomography,”Sci.Rep.10(1),19513(2020).

[0007] [3]N.Lippok,M.Villiger,and B.E.Bouma,“Degree of polarization(uniformity)and depolarization index:unambiguous depolarization contrast foroptical coherence tomography,”Opt.Lett.40(17),3954-3957(2015).

[0008] [4]N.Lippok,B.Braaf,M.Villiger,et al.,“Quantitative depolarizationmeasurements for fiber-based polarization-sensitive optical frequency domainimaging of the retinal pigment epithelium,”J.Biophotonics 12(1),e201800156(2019).

[0009] [5]M.Yamanari,S.Tsuda,T.Kokubun,et al.,“Estimation of Jones matrix,birefringence and entropy using Cloude-Pottier decomposition in polarization-sensitive optical coherence tomography,”Biomed.Opt.Express 7(9),3551-3573(2016).

[0010] [6] Z.Ding, C.-P.Liang, and Y.Chen, "Technology developments and biomedical applications of polarization-sensitive optical coherencetomography," Front.Optoelectron.8(2),128-140(2015). Summary of the Invention

[0011] The technical problem to be solved by this invention is to provide a precise medium depolarization measurement method for polarization-sensitive coherence tomography of intravascular catheters. The technical solution adopted by this invention is:

[0012] A method for depolarization measurement of intravascular polarization-sensitive optical coherence tomography media, used in a catheter polarization-sensitive optical coherence tomography system, includes the following steps:

[0013] The first step is to set the polarization state of the input light to the catheter polarization-sensitive optical coherence tomography system as E. in The reference light is represented as E. ref Set the reference light in the H and V channels of the input light and reference light in the system to have the same intensity.

[0014] The second step involves acquiring the electrical signals measured at the polarization diversity points in the form of a Jones matrix, performing dispersion compensation and interpolation Fourier transform to generate a spatial image, and then performing image segmentation. A mean window with dimensions of 2p and 2q pixels is constructed centered at pixel (i,j). The Jones matrices of each point within the window and the Jones matrices at the selected reference plane location are converted into the mean Mueller matrix using matrix transformation and matrix averaging algorithms, respectively, to obtain the mean measurement Mueller matrix M. ij , where m ij (u,v)u,v∈{1,2,3,4} is the average measurement Mueller matrix M. ij The elements are obtained by averaging the elements at corresponding positions in the Mueller matrix measured within the window:

[0015]

[0016] The third step is to select the image region of the outer wall of the catheter in the scan data as the reference region, and denote the number of pixels in the region as N. ROR The average measured Mueller matrix per pixel within the marked reference area is M. ij-ROR The average measurement Mueller matrix within the reference area is calculated as M. ij-ROR average And thus calculate the standard deviation matrix S of the reference region;

[0017]

[0018] The fourth step is to apply the Lu-Chipman matrix decomposition to the average measurement Mueller matrix over the reference region, which is M. ij-ROR Obtain the debiased Mueller matrix M in the reference region ij-ROR △ ;

[0019] (1) Extract the average measurement Mueller matrix M of the reference area ij-ROR The bottom right corner contains a block matrix m with nine elements. ij-ROR :

[0020]

[0021] (2) Calculate the average measurement Mueller matrix M of the reference area. ij-ROR block vector P ij-ROR D ij-ROR and take D ij-ROR unit vector

[0022]

[0023]

[0024]

[0025] (3) Calculate the average double-attenuation block matrix m of the reference region ij-ROR-D :

[0026]

[0027] Where E x Let x be the identity matrix containing x×x elements;

[0028] (4) Construct the reference region averaged double attenuation matrix M ij-ROR D :

[0029]

[0030] (5) Use matrix inverse operation to transform the average double decay matrix M ij-ROR D The Mueller matrix M is measured from the average of the reference region. ij-ROR After removing the middle, the pseudo-average birefringence phase delay matrix M of the reference region is obtained. ij-ROR Rf :

[0031] M ij-ROR Rf =M ij-ROR (M ij-ROR D) -1 (9)

[0032] (6) The pseudo-average birefringence phase delay matrix M of the reference region ij-ROR Rf Block representation and extraction of block matrix m ij-ROR 'and vector P ij-ROR-△ :

[0033]

[0034] (7) Applying the matrix decomposition rule to obtain the pseudo-average birefringence phase delay matrix M in the reference region ij-ROR Rf Extract the average debiasing matrix M of the reference region ij-ROI △ ;

[0035] Let λ a , λ b , λ c For m ij-ROR '(m ij-ROR ') T If the three eigenvalues ​​are such that the bias-reduction block matrix m is obtained, then... ij-ROR-△ Represented as:

[0036]

[0037] (8) Reference region average debiasing matrix M ij-ROR △ This can be represented in the following block-based form:

[0038]

[0039] Fifth step: Calculate the average debiasing matrix M of the reference region. ij-ROR △ The Frobenius norm ||M ij-ROR △ || F The Frobenius norm ||S|| of the standard deviation matrix of the reference region F Comparison:

[0040] (1) Set error:

[0041] (2) If:

[0042] ||M ij-ROR △ || F ≤||S|| F +error (13)

[0043] The debiasing caused by noise at position (i, j) is negligible;

[0044] (3) If:

[0045] ||M ij-ROR △ || F >||S|| F +error (14)

[0046] Then there is noise-induced debiasing at position (i, j);

[0047] Step 6: Apply Lu-Chipman matrix decomposition to the target region to average the measurement Mueller matrix M. ij-ROI Obtain the debiased Mueller matrix M of the target region ij-ROI △ ;

[0048] (1) Extract the average measurement Mueller matrix M of the target area ij-ROI The bottom right corner contains a block matrix m with nine elements. ij-ROI :

[0049]

[0050] (2) Calculate the average measurement Mueller matrix M of the target area. ij-ROI block vector P ij-ROI D ij-ROI and take D ij-ROI unit vector

[0051]

[0052]

[0053]

[0054] (3) Calculate the average double-attenuation block matrix m of the target region. ij-ROI-D :

[0055]

[0056] (4) Construct the average double decay matrix M of the target region ij-ROI D :

[0057]

[0058] (5) Use matrix inverse operation to calculate the average double attenuation matrix M of the target region. ij-ROI D The average measurement of the Mueller matrix M from the target region ij-ROI After removing the middle, the pseudo-average birefringence phase delay matrix M of the target region is obtained. ij-ROI Rf:

[0059] M ij-ROI Rf =M ij-ROI (M ij-ROI D ) -1 (twenty one)

[0060] (6) The pseudo-average birefringence phase delay matrix M of the target region ij-ROI Rf Block representation and extraction of block matrix m ij-ROI 'and vector P ij-ROI-△ :

[0061]

[0062] (7) Apply the matrix decomposition rule to obtain the pseudo-average birefringence phase delay matrix M of the target region. ij-ROI Rf Further decomposed into the average birefringence phase delay matrix M of the target region ij-ROI R and the target region average debiasing matrix M ij-ROI △ ;

[0063] Let λ1, λ2, λ3 be m ij-ROI '(m ij-ROI ') T If the three eigenvalues ​​are such that the bias-reduction block matrix m is obtained, then... ij-ROI-△ Represented as:

[0064]

[0065] (8) Average the debiasing matrix M of the target region ij-ROI △ The blocks are divided into the following forms:

[0066]

[0067] (9) Obtain the average birefringence phase delay matrix M of the target region through inverse matrix operation. ij-ROI R :

[0068] M ij-ROI R =(M ij-ROI △ ) -1 M ij-ROI Rf (25)

[0069] Step 7: Calculate the average debiasing matrix M for the target region. ij-out △ The Frobenius norm ||Mij-out △ || F The Frobenius norm ||S|| of the standard deviation matrix of the reference region F Comparison:

[0070] (1) Set error2:

[0071] (2) If:

[0072] ||M ij-ROI △ || F ≤||S|| F +error2 (26)

[0073] The debiasing caused by noise at position (i, j) is negligible;

[0074] (3) If:

[0075] ||M ij-ROI △ || F >||S|| F +error2 (27)

[0076] Then there is noise-induced debiasing at position (i, j);

[0077] Step 8: Perform noise compensation on the positions that satisfy the relationship (27) and select the average debiasing matrix M at that position. ij-ROI △ Set the test coefficients α and construct the test average debiasing matrix M. ij-test △ (α):

[0078]

[0079] Step 9: Construct the test measurement Mueller matrix M ij-ROI-test (α)

[0080] M ij-ROI-test (α)=M ij-ROI △ (α)M ij-ROI R M ij-ROI D (29)

[0081] Step 10: Construct the trial measurement Mueller matrix M ij-ROI-trial (α)

[0082] M ij-ROI-trial (α)=M ij-test △ (α)M ij-ROI RM ij-ROI D (30)

[0083] Step 11: Gradually decrease the coefficients from 1 to 0, substituting different test coefficients α, until the following relationship is satisfied.

[0084] ||M ij-ROI △ (M ij-test △ (α)) -1 || F >||S F +error (31)

[0085] Step 12, let M be the Mueller matrix of the trial measurement at this time. ij-ROI-trial αk And based on this, the medium depolarization coefficient EMDI is calculated:

[0086]

[0087] This method is based on the Mueller matrix decomposition method for determining depolarization properties, combined with the moving window averaging Mueller matrix algorithm, and incorporates an iterative approximation method for depolarization matrix to achieve high-precision estimation of the medium's depolarization index for depolarization imaging. This significantly removes the influence of noise on depolarization measurements, improves the quality of depolarized images, and increases the probability of identifying highly depolarized pathological patches. Compared with depolarization measurement methods based on ordinary depolarization coefficients and polarization homogenization, the depolarization measurement method applied here can effectively remove the influence of noise in polarization-sensitive coherence tomography with dual polarization states, exhibiting wider applicability and higher robustness. Attached Figure Description

[0088] Figure 1 This is a schematic diagram of the structure of the catheter polarization-sensitive optical coherence tomography system of the present invention;

[0089] Figure 2 This is a flowchart of a method for depolarization measurement of polarization-sensitive coherence tomography media for intravascular catheters according to the present invention.

[0090] Figure 3 Image of a porcine coronary atherosclerotic cholesterol plaque phantom. The high-density depolarization plaque is placed in the lower left of the image. Left: Intensity image; Left-middle: High noise resistance depolarization index image calculated by the algorithm; Right-middle: Depolarization index image without noise compensation; Right: Polarization uniformity image without noise compensation. The colors from bright to dark indicate the degree of depolarization influence from large to small. The white arrow points to the location of the real plaque, and the gray arrow points to the location of the false plaque introduced by noise.

[0091] Figure 4Image of a 1:1 ratio cholesterol-protein plaque phantom in porcine coronary atherosclerosis. The highly depolarized plaque is placed in the lower left of the image. Left: Intensity image; Left-middle: High noise resistance depolarization index image calculated by the algorithm; Right-middle: Depolarization index image without noise compensation; Right: Polarization uniformity image without noise compensation. The colors from bright to dark indicate the degree of depolarization influence from large to small. White arrows point to the location of the real plaque, and gray arrows point to the location of the false plaque introduced by noise.

[0092] Figure 5 Image of the porcine coronary artery, without high depolarization plaques. Left: Intensity image; Left middle: High noise resistance depolarization index image calculated by the algorithm; Right middle: Depolarization index image without noise compensation; Right: Polarization uniformity image without noise compensation. The colors from bright to dark indicate the degree of influence of depolarization from large to small. The gray arrows point to the location of false plaques introduced by noise. Detailed Implementation

[0093] This invention relates to a method for measuring media depolarization in intravascular polarization-sensitive optical coherence tomography (PS-OCT). Specifically, it addresses how to reduce the interference of noise depolarization on the imaging of media depolarization coefficients in catheter-based PS-OCT images. This method eliminates the influence of noise depolarization caused by the stretching and compression of the optical fiber by the catheter motor during high-speed rotation, enhancing the clarity of the depolarization coefficient image, increasing the probability of identifying atherosclerotic plaques, and improving the imaging quality of PS-OCT depolarization coefficients. This invention enables the PS-OCT system to separate the sample's inherent depolarization properties—namely, media depolarization and noise-introduced noise depolarization—from the depolarization measurement values, extracting and accurately calculating them separately. This achieves media depolarization imaging, obtaining depolarization measurement values ​​that are closer to the sample's media depolarization than other PS-OCT depolarization imaging techniques. It expands and optimizes intravascular catheter OCT detection parameters, improving the ability to analyze microscopic lesions within blood vessels. This invention utilizes the Jones matrix to extract polarization information and then converts it to the average Mueller matrix for polarization feature characterization. Depolarization is extracted through matrix incoherence averaging. The presence and impact of noise are evaluated using Lu-Chipman matrix decomposition, Frobenius norm discrimination, and iterative approximation of the test matrix. Noise compensation is then applied to the depolarization matrix. Depolarization matrices that meet the noise-free criteria are classified as noise-free depolarization matrices, which can be used to express medium depolarization. The depolarization index is calculated using these matrices to obtain the accurate medium depolarization coefficient distribution of the sample, achieving precise measurement of PS-OCT medium depolarization in the catheter.

[0094] The following detailed description of a method for depolarization measurement of polarization-sensitive coherence tomography media for intravascular catheters, in conjunction with embodiments and accompanying drawings, provides a clearer picture of the present invention.

[0095] This invention provides a method for depolarization measurement of polarization media in duct polarization-sensitive optical coherence tomography, used for, for example... Figure 1 The working principle of the catheter polarization-sensitive optical coherence tomography (PS-OCT) system shown is as follows:

[0096] The outgoing light from the scanning light source 1 of the guide tube PS-OCT system enters through port 1 of the 1:99 fiber coupler 2 and is distributed to the sample arm and reference arm respectively from ports 2 and 3 at a ratio of 1:99. The outgoing light from port 2 of the 1:99 fiber coupler 2 enters the sample arm, and the beam entering the sample arm is incident on the three-ring polarization controller 3 and then on the 18.5-meter-long polarization-maintaining fiber 4, entering port 1 of the circulator 6. The light exits from port 2 of the circulator 6, passes through the rotation mechanism 8 and enters the imaging guide tube 11. After being reflected by the sample, the light returns from the imaging guide tube 11 to the circulator 6 and exits through port 3 of the circulator 6. The outgoing light from port 3 of the 1:99 fiber coupler 2 enters the reference arm, and the light entering the reference arm is incident on the 18.5-meter-long single-mode fiber 5. The outgoing light enters port 1 of the circulator 7, exits from port 2 and enters the reflective fiber delay line 10. The reflected light enters through port 2 of the circulator 7 and exits from port 3 to the three-ring polarization controller 9. The light emitted from port 3 of the sample arm via circulator 6 and the light emitted from the reference arm via triple-ring polarization controller 9 are incident on ports 1 and 2 of fiber coupler 12 respectively, and interfere with each other. They then enter triple-ring polarization controller 13 and triple-ring polarization controller 14 from ports 3 and 4 respectively, with a ratio of 50:50. The emitted light is incident on polarization beamsplitters 15 and 16 respectively. The emitted light from fiber beamsplitter 15 is incident on balance detectors 17 and 18 from ports 1 and 2 respectively. The emitted light from polarization beamsplitter 16 is incident on balance detectors 17 and 18 from ports 1 and 2 respectively. The electrical signals from balance detectors 17 and 18 are received by acquisition card 19 and transmitted to computer 20.

[0097] The system employs a fast-scanning light source and utilizes polarization-maintaining fiber to generate orthogonal polarization state delays. Polarization diversity acquisition is performed via a polarization beam splitter. The length of the polarization-maintaining fiber depends on its birefringence, resulting in a phase delay equal to half the imaging depth of ordinary OCT. This method ensures that the system can simultaneously present polarization-divided imaging of two orthogonal input polarization states in a single image, providing a basis for subsequent total depolarization and depolarization classification during extraction experiments.

[0098] like Figure 2 As shown, the present invention provides a method for depolarization measurement of intravascular polarization-sensitive optical coherence tomography media, comprising the following steps:

[0099] The first step is to set the polarization state of the input light to the catheter polarization-sensitive optical coherence tomography (PS-OCT) system as E. inThe reference light is represented as E. ref Set the reference light in the H and V channels of the input light and reference light in the system to have the same intensity.

[0100] The second step is to acquire the electrical signal measured at the polarization diversity point in the form of a Jones matrix, and perform dispersion compensation, interpolation, and Fourier transform to generate a spatial domain image [6], and then perform image segmentation; with the sample depth position j and the horizontal distance i in the image as the center, construct an average window with a length and width of 2p and 2q pixels respectively, and convert the Jones matrix of each point in the window and the Jones matrix at the selected reference surface position into the average Mueller matrix through matrix transformation and matrix averaging algorithms respectively, to obtain the average measurement Mueller matrix M. ij , where m ij (u,v)u,v∈{1,2,3,4} is the average measurement Mueller matrix M. ij The elements are obtained by averaging the elements at corresponding positions in the Mueller matrix measured within the window:

[0101]

[0102] The third step is to select the image region of the outer wall of the catheter in the scan data as the reference region, and denote the number of pixels in the region as N. ROR The average measured Mueller matrix per pixel within the marked reference area is M. ij-ROR The average measurement Mueller matrix within the reference area is calculated as M. ij-ROR average And from this, the standard deviation matrix S of the reference region is calculated.

[0103]

[0104] The fourth step is to apply the Lu-Chipman matrix decomposition to the average measurement Mueller matrix over the reference region, which is M. ij-ROR Obtain the debiased Mueller matrix M in the reference region ij-ROR △ .

[0105] (1) Extract the average measurement Mueller matrix M of the reference area ij-ROR The bottom right corner contains a block matrix m with nine elements. ij-ROR :

[0106]

[0107] (2) Calculate the average measurement Mueller matrix M of the reference area. ij-ROR block vector P ij-ROR D ij-ROR and take D ij-ROR unit vector

[0108]

[0109]

[0110]

[0111] (3) Calculate the average double-attenuation block matrix m of the reference region ij-ROR-D :

[0112]

[0113] Where E x Let x be the identity matrix containing x×x elements.

[0114] (4) Construct the reference region averaged double attenuation matrix M ij-ROR D :

[0115]

[0116] (5) Use matrix inverse operation to transform the average double decay matrix M ij-ROR D The Mueller matrix M is measured from the average of the reference region. ij-ROR After removing the middle, the pseudo-average birefringence phase delay matrix M of the reference region is obtained. ij-ROR Rf :

[0117] M ij-ROR Rf =M ij-ROR (M ij-ROR D ) -1 (9)

[0118] (6) The pseudo-average birefringence phase delay matrix M of the reference region ij-ROR Rf Block representation and extraction of block matrix m ij-ROR 'and vector P ij-ROR-△ :

[0119]

[0120] (7) Applying the matrix decomposition rule to obtain the pseudo-average birefringence phase delay matrix M in the reference region ij-ROR Rf Extract the average debiasing matrix M of the reference region ij-ROI △ ;

[0121] Let λ a , λ b , λ c For m ij-ROR '(m ij-ROR ') TIf the three eigenvalues ​​are such that the bias-reduction block matrix m is obtained, then... ij-ROR-△ It can be represented as:

[0122]

[0123] (8) Reference region average debiasing matrix M ij-ROR △ This can be represented in the following block-based form:

[0124]

[0125] Fifth step: Calculate the average debiasing matrix M of the reference region. ij-ROR △ The Frobenius norm ||M ij-ROR △ || F The Frobenius norm ||S|| of the standard deviation matrix of the reference region F Comparison:

[0126] (1) Set the error according to the experimental requirements:

[0127] (2) If:

[0128] ||M ij-ROR △ || F ≤||S|| F +error (13)

[0129] The debiasing caused by noise at position (i, j) can be ignored.

[0130] (3) If:

[0131] ||M ij-ROR △ || F >||S|| F +error (14)

[0132] Then there is noise at position (i, j) causing debiasing.

[0133] Step 6: Apply Lu-Chipman matrix decomposition to the target region to average the measurement Mueller matrix M. ij-ROI Obtain the debiased Mueller matrix M of the target region ij-ROI △ .

[0134] (1) Extract the average measurement Mueller matrix M of the target area ij-ROI The bottom right corner contains a block matrix m with nine elements. ij-ROI :

[0135]

[0136] (2) Calculate the average measurement Mueller matrix M of the target area. ij-ROI block vector P ij-ROI D ij-ROI and take D ij-ROI unit vector

[0137]

[0138]

[0139]

[0140] (3) Calculate the average double-attenuation block matrix m of the target region. ij-ROI-D :

[0141]

[0142] (4) Construct the average double decay matrix M of the target region ij-ROI D :

[0143]

[0144] (5) Use matrix inverse operation to obtain the average birefringence phase delay matrix M of the target region. ij-ROI D The average measurement of the Mueller matrix M from the target region ij-ROI After removing the middle, the pseudo-average birefringence phase delay matrix M of the target region is obtained. ij-ROI Rf :

[0145] M ij-ROI Rf =M ij-ROI (M ij-ROI D ) -1 (twenty one)

[0146] (6) The pseudo-average birefringence phase delay matrix M of the target region ij-ROI Rf Block representation and extraction of block matrix m ij-ROI 'and vector P ij-ROI-△ :

[0147]

[0148] (7) Apply the matrix decomposition rule to obtain the pseudo-average birefringence phase delay matrix M of the target region. ij-ROI Rf Further decomposed into the average birefringence phase delay matrix M of the target region ij-ROIR and the target region average debiasing matrix M ij-ROI △ First, let λ1, λ2, and λ3 be m. ij-ROI '(m ij-ROI ') T If the three eigenvalues ​​are such that the bias-reduction block matrix m is obtained, then... ij-ROI-△ It can be represented as:

[0149]

[0150] (8) Average the debiasing matrix M of the target region ij-ROI △ The blocks are divided into the following forms:

[0151]

[0152] (9) Obtain the average birefringence phase delay matrix M of the target region through inverse matrix operation. ij-ROI R :

[0153] M ij-ROI R =(M ij-ROI △ ) -1 M ij-ROI Rf (25)

[0154] Step 7: Calculate the average debiasing matrix M for the target region. ij-out △ The Frobenius norm ||M ij-out △ || F The Frobenius norm ||S|| of the standard deviation matrix of the reference region F Comparison:

[0155] (1) Set the error according to the experimental requirements:

[0156] (2) If:

[0157] ||M ij-ROI △ || F ≤||S|| F +error (26)

[0158] The debiasing caused by noise at position (i, j) can be ignored.

[0159] (3) If:

[0160] ||M ij-ROI △ || F >||S||F +error (27)

[0161] Then there is noise at position (i, j) causing debiasing.

[0162] Step 8: Perform noise compensation on the positions that satisfy the relationship (28) and select the average debiasing matrix M at that position. ij-ROI △ Set the test coefficients α and construct the test average debiasing matrix M. ij-test △ (α):

[0163]

[0164] Step 9: Construct the test measurement Mueller matrix M ij-ROI-test (α)

[0165] M ij-ROI-test (α)=M ij-ROI △ (α)M ij-ROI R M ij-ROI D (29)

[0166] Step 10: Construct the trial measurement Mueller matrix M ij-ROI-trial (α)

[0167] M ij-ROI-trial (α)=M ij-test △ (α)M ij-ROI R M ij-ROI D (30)

[0168] Step 11: Gradually decrease the coefficients from 1 to 0, substituting different test coefficients α, until the following relationship is satisfied.

[0169] ||M ij-ROI △ (M ij-test △ (α)) -1 || F >||S F +error (31)

[0170] Step 12, let M be the Mueller matrix of the trial measurement at this time. ij-ROI-trial αk And based on this, the medium depolarization coefficient EMDI is calculated:

[0171]

[0172] Step 13: Apply the method shown in steps 1 to 12 to perform traversal calculations on the phase delay at each position in polar coordinates to obtain the two-dimensional polar coordinate distribution map of the depolarized medium.

[0173] Step fourteen involves performing coordinate interpolation transformation on the phase delay calculation results shown in step seven, converting polar coordinates to Cartesian coordinates, and finally obtaining the depolarized image of the sample from the duct polarization-sensitive optical coherence tomography system.

[0174] The coordinate interpolation transformation is necessary because during the data acquisition process of the PS-OCT system, the depth information A-Scan and the lateral information B-Scan are imaged, and the final image output is a polar coordinate image. However, the actual requirement is an image inside the lumen, so the processed polar coordinate image needs to be processed into a PS-OCT image in Cartesian coordinates.

[0175] like Figure 3-5 The image shown is an illustration of the high-noise-resistance depolarization measurement method for intravascular catheter polarization-sensitive coherence tomography used in this invention. Left: Intensity image; Left-middle: High-noise-resistance depolarization index image calculated by the algorithm; Right-middle: Depolarization index image without noise compensation; Right: Polarization homogenization image without noise compensation. Colors from red to blue indicate the degree of depolarization influence from high to low, and light to dark indicate signal intensity from high to low. White arrows point to the location of the true plaque, and yellow arrows point to the location of spurious plaques introduced by noise. This method effectively removes spurious high-depolarization plaques introduced by noise.

[0176] Although the present invention has been described above in conjunction with the accompanying drawings, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many modifications under the guidance of the present invention without departing from the spirit of the present invention, and these modifications are all within the protection scope of the present invention.

Claims

1. A method for depolarization measurement of intravascular polarization-sensitive optical coherence tomography media, used in a catheter polarization-sensitive optical coherence tomography system, comprising the following steps: The first step is to set the polarization state of the input light to the catheter polarization-sensitive optical coherence tomography system as follows: E in ， The reference light is represented as E ref Set the reference light in the H and V channels of the input light and reference light in the system to have the same intensity. The second step involves acquiring the electrical signals measured at the polarization diversity point in the form of a Jones matrix, performing dispersion compensation and interpolation Fourier transform to generate a spatial domain image, and then performing image segmentation. Centered on pixel (i,j), an averaging window is constructed. The Jones matrix of each point within the window and the Jones matrix at the selected reference surface position are converted into the average Mueller matrix through matrix transformation and matrix averaging algorithms, respectively, to obtain the average measurement Mueller matrix M. ij ,in For the average measurement Mueller matrix M ij The elements are obtained by averaging the elements at corresponding positions in the Mueller matrix measured within the window: （1） The third step is to select the image region of the outer wall of the catheter in the scan data as the reference region, and denote the number of pixels in the region as... N ROR The average measured Mueller matrix per pixel within the marked reference area is M. ij-ROR The average measurement Mueller matrix within the reference area is calculated as M. ij-ROR average And thus calculate the standard deviation matrix S of the reference region; （2） The fourth step is to apply the Lu-Chipman matrix decomposition to the average measurement Mueller matrix over the reference region, which is M. ij-ROR Obtain the debiased Mueller matrix of the reference region ; (1) Extract the average measurement Mueller matrix M of the reference area ij-ROR The bottom right corner contains a block matrix with nine elements. : （3） (2) Calculate the average measurement Mueller matrix M of the reference area ij-ROR block vector , and take unit vector : （4） （5） （6） (3) Calculate the average double-attenuation block matrix of the reference region : （7） Where E x For inclusion x × x An identity matrix with n elements; (4) Construct the reference region average double attenuation matrix : （8） (5) Use matrix inverse operation to transform the average double decay matrix The Mueller matrix M is measured from the average of the reference region. ij-ROR After removing the middle, the pseudo-average birefringence phase delay matrix of the reference region is obtained. : （9） (6) The pseudo-average birefringence phase delay matrix of the reference region Block representation and extraction of block matrix and vector : （10） (7) Apply the matrix decomposition rule to obtain the pseudo-average birefringence phase delay matrix of the reference region. Extract the average debiasing matrix of the reference region ; set up , , for The three eigenvalues ​​are then the bias-reducing block matrix. Represented as: （11） (8) Reference region average debiasing matrix This can be represented in the following block-based form: （12） Step 5: Calculate the average debiasing matrix for the reference region. Frobenius norm Frobenius norm of the standard deviation matrix of the reference region Comparison: (1) Set error: (2) If: （13） The debiasing caused by noise at position (i, j) is negligible; (3) If: （14） Then there is noise-induced debiasing at position (i, j); Step 6: Apply Lu-Chipman matrix decomposition to the target region to average the measurement Mueller matrix M. ij-ROI Obtain the debiased Mueller matrix of the target region ; (1) Extract the average measurement Mueller matrix M of the target area ij-ROI The bottom right corner contains a block matrix with nine elements. : （15） (2) Calculate the average measurement Mueller matrix M of the target area ij-ROI block vector , and take unit vector : （16） （17） （18） (3) Calculate the average double-attenuation block matrix of the target region : （19） (4) Construct the average double decay matrix of the target region : （20） (5) Use matrix inverse operation to average the double attenuation matrix of the target region. The average measurement of the Mueller matrix M from the target region ij-ROI After removing the middle, the pseudo-average birefringence phase delay matrix of the target region is obtained. : （21） (6) The pseudo-average birefringence phase delay matrix of the target region Block representation and extraction of block matrix and vector : （22） (7) Apply the matrix decomposition rule to determine the pseudo-average birefringence phase delay matrix of the target region. Further decomposed into the average birefringence phase delay matrix of the target region and the average debiasing matrix of the target region ; set up , , for The three eigenvalues ​​are then the bias-reducing block matrix. Represented as: （23） (8) Average the debiasing matrix of the target region The blocks are divided into the following forms: （24） (9) Obtain the average birefringence phase delay matrix of the target region through inverse matrix operation. : （25） Step 7: Calculate the average debiasing matrix for the target region. Frobenius norm Frobenius norm of the standard deviation matrix of the reference region Comparison: (1) Set error2: (2) If: （26） The debiasing caused by noise at position (i, j) is negligible; (3) If: （27） Then there is noise-induced debiasing at position (i, j); Step 8: Perform noise compensation on the positions that satisfy the relationship (27) and select the average debiasing matrix for that position. Set the test coefficient α and construct the test average debiasing matrix. : （28） Step 9: Construct the test measurement Mueller matrix M ij-ROI-test (α) （29） Step 10: Construct the trial measurement Mueller matrix M ij-ROI-trial (α) （30） Step 11: Gradually decrease the coefficients from 1 to 0, substituting different test coefficients α, until the following relationship is satisfied. （31） Step 12, let M be the Mueller matrix of the trial measurement at this time. ij-ROI-trial αk And based on this, the medium depolarization coefficient EMDI is calculated: （32）。

Images