A catheter polarization-sensitive optical coherence tomography system and demodulation method

Abstract

A catheter polarization-sensitive optical coherence tomography system and demodulation method, characterized in that the light source in the system adopts a fast scanning light source, the system adopts a polarization-maintaining optical fiber to generate a delay of an orthogonal polarization state, and polarization diversity acquisition is performed through a polarization beam splitter, and polarization diversity imaging of two orthogonal input polarization states is simultaneously presented in one image. In the demodulation method, a series of steps such as polarization leveling, background signal elimination, spectrum shaping, two-state dispersion elimination, interpolation Fourier transform, reference plane selection, polarization solution, and polar coordinate conversion to Cartesian coordinates are finally implemented to achieve vascular birefringence imaging. The present invention can obtain vascular birefringence information and provide more characteristic information for cardiovascular and cerebrovascular lesions.

Description

Technical Field

[0001] The present invention relates to catheter optical coherence tomography (OCT) covering the field of catheter imaging such as cardiovascular and cerebrovascular diseases, and in particular to birefringence information of demodulated samples in catheter polarization-sensitive OCT (PS-OCT) images, specifically a catheter polarization-sensitive optical coherence tomography system and demodulation method used in OCT technology. Background Art

[0002] Catheter OCT imaging technology is currently the vascular imaging method with the highest image resolution, especially catheter PS-OCT imaging technology, which can solve the medical problem of difficult in vivo, real-time and rapid judgment of the stability of atherosclerotic plaques, and can improve the prevention and treatment of atherosclerotic diseases. However, the existing OCT system has reached a level of resolution that can determine the nature of tissue plaques, but it is still insufficient in tissue penetration, image clarity and accuracy of tissue plaque type judgment. Using PS-OCT technology and improving related technical performance is the key direction of the development of OCT systems, and is also the only way to solve the aforementioned key scientific problems.

[0003] In terms of catheter OCT, catheter PS-OCT is an extension of catheter OCT technology, which provides a quantitative measurement method for the birefringence properties of tissues. The birefringence of light changes the polarization state of light and can be associated with proteins and biomacromolecules with oriented structures such as collagen and actin. The birefringence phenomenon enhanced by catheter PS-OCT is closely related to the presence of a large number of thick collagen fibers or smooth muscle cells in the intima, so the high-resolution detection of catheter PS-OCT imaging can be applied to the measurement of enhanced plaque stability. In addition, the catheter PS-OCT system has the potential to evaluate plaque collagen and distinguish between normal intima, fibrous plaques, lipid plaques, and calcified plaques. Summary of the invention

[0004] The purpose of the present invention is to solve the problem that the existing OCT catheter optical coherence tomography system can only obtain vascular intensity information, and to invent a catheter polarization-sensitive optical coherence tomography system and demodulation method to obtain vascular birefringence information and provide more characteristic information for cardiovascular and cerebrovascular diseases. In the system, a fast scanning light source is used as the light source, and a polarization-maintaining optical fiber is used in the system to generate delays in orthogonal polarization states. Polarization diversity acquisition is performed through a polarization beam splitter, and polarization diversity imaging of two orthogonal input polarization states is simultaneously presented in one image. In terms of the demodulation method, a series of steps such as polarization leveling, background signal elimination, spectrum shaping, dual-state dispersion elimination, interpolation Fourier transform, reference plane selection, polarization solution, and conversion of polar coordinates to Cartesian coordinates are used to finally achieve vascular birefringence imaging.

[0005] One of the technical solutions of the present invention is:

[0006] A catheter polarization-sensitive optical coherence tomography system comprises a scanning light source 1, wherein the light emitted from the scanning light source 1 enters from port 1 of a fiber coupler 2 and is distributed to a sample arm and a reference arm from ports 22 and 3 of the fiber coupler respectively; the light emitted from port 2 of the fiber coupler 2 enters the sample arm, the light beam entering the sample arm is incident on a polarization (three-ring) controller 3 and then incident on a polarization-maintaining optical fiber 4 and enters port 1 of a first circulator 6, the light is emitted from port 2 of the first circulator 6, the emitted light enters an imaging catheter (11) through a rotating mechanism (8), the light reflected by the sample returns from the imaging catheter (11) to the first circulator 6 and is emitted through port 3 of the first circulator 6; the light emitted from port 3 of the fiber coupler 2 enters the reference arm, the light entering the reference arm is incident on a single-mode optical fiber 5, the emitted light from the single-mode optical fiber 5 enters port 1 of a second circulator 7, and is emitted from port 2 of the second circulator 7 and enters a reflective optical fiber delay line 10, the reflective optical fiber delay line 10 is reflected from the sample. The reflected light of the optical fiber delay line 10 is incident through port 2 of the second circulator 7, and is emitted from port 3 of the second circulator 7 to the first polarization controller 9. The output light of the sample arm through port 3 of the first circulator 6 and the output light of the reference arm through the first polarization controller 9 are respectively incident on the optical fiber coupler 12 from ports 1 and 2 of the optical fiber coupler 12 to interfere, and enter the second polarization controller 13 and the fourth polarization controller 14 from ports 3 and 4 of the optical fiber coupler 12 respectively. The output light is respectively incident on the first polarization beam splitter 15 and the second polarization beam splitter 16. The output light of the first polarization beam splitter 15 is respectively incident on the first balanced detector 17 and the second balanced detector 18 from ports 1 and 2 thereof. The output light of the second polarization beam splitter 16 is respectively incident on the first balanced detector 17 and the second balanced detector 18 from ports 1 and 2 thereof. The electrical signals of the first balanced detector 17 and the second balanced detector 18 are received by the acquisition card 19 and transmitted to the computer 20.

[0007] The scanning light source 1 is a fast scanning light source, which uses polarization-maintaining fiber to generate delays in orthogonal polarization states, and performs polarization diversity acquisition through a polarization beam splitter. The length of the polarization-maintaining fiber depends on the phase delay generated by its birefringence, which is equal to half of the OCT imaging depth, to ensure that polarization diversity imaging of two orthogonal input polarization states is simultaneously presented in one image, making it possible to subsequently eliminate the system birefringence changes introduced by catheter rotation.

[0008] The optical fiber coupler 2 is a 1:99 optical fiber coupler.

[0009] The outgoing light of the scanning light source 1 enters from port 1 of the 1:99 optical fiber coupler 2, and is distributed to the sample arm and the reference arm from ports 2 and 3 of the optical fiber coupler 2 at a ratio of 1:99.

[0010] The lengths of the polarization-maintaining optical fiber 4 and the single-mode optical fiber 5 are both 18.5 meters.

[0011] The fiber coupler 12 is a 50:50 fiber coupler; the outgoing light of the sample arm passing through the 3-port of the first circulator 6 and the outgoing light of the reference arm passing through the polarization controller 9 are respectively incident on the fiber coupler 12 from the 1-port and 2-port of the fiber coupler 12 to interfere, and enter the first polarization controller 13 and the second polarization controller 14 from the 3-port and 4-port of the fiber coupler 12 at a ratio of 50:50.

[0012] The second technical solution of the present invention is:

[0013] A catheter polarization-sensitive optical coherence tomography demodulation method is characterized in that it comprises the following steps:

[0014] Step 1: polarization leveling: on the one hand, the light intensity of the reference light in the H path and the V path in the polarization concentration is equal; on the other hand, the light of the sample arm is evenly distributed to the two polarization controllers 13 and 14;

[0015] Step 2: Background signal removal. Background noise and baseband seriously affect the imaging quality, especially baseband imaging. The background noise and baseband need to be removed.

[0016] Step 3: Dual-state dispersion compensation, process the data collected by the PS-OCT system, perform Hilbert transform on each A-Line of the PS-OCT original image of channel H and channel V in polar coordinates, express the PS-OCT interference signal in the form of complex exponentials, and then perform numerical dispersion compensation on each A-Line after Hilbert transform, calculate the correction phase Φ(k) through the second-order and third-order dispersion coefficients, and multiply it with the original phase of the A-Line in the form of e-exponential, and realize numerical dispersion compensation of the original signal of the A-Line by phase addition and subtraction, and correct the frequency broadening of the interference signal;

[0017] Step 4: Spectral shaping: To prevent spectrum leakage of the range domain signal after Fourier transform, a cosine tapered window is used to shape the spectrum of the A-Line data.

[0018] Step 5: Perform interpolation Fourier transform. Due to depth multiplexing, the number of data points in the depth direction is small, and the interpolation Fourier transform is needed to fill the number of points to make it consistent with the number of points on the circumference. Finally, the two-dimensional data information of PS-OCT is obtained, that is, the distance domain information H1+H2 and V1+V2 of the dual-state superposition of channel H and channel V;

[0019] Step 6: Two-state image segmentation, segment the distance domain information H1+H2 and V1+V2 of the two-state superposition of channel H and channel V, perform cross-correlation operation on the two segments of each column of A-line data after selecting the segmentation point, adjust the segmentation point by the cross-correlation peak frequency shift value, and finally obtain four groups of data H1, H2, V1 and V2;

[0020] Step 7: Before performing polarization solution, select the reference position, that is, the position of the zero point of the birefringence phase delay of each column of A-line data. The outer surface of the catheter sheath is selected as the reference surface, which is not affected by the sample to be tested. The position of the outer surface of the catheter sheath is selected using a data frame, and the points with the largest intensity in each column of A-line data falling in the data frame are used as reference points.

[0021] Step 8: Substitute each column of A-line data of H1, H2, V1 and V2 into the polarization solution algorithm to obtain the birefringence phase delay;

[0022] Step nine: Perform coordinate interpolation transformation on the birefringence phase delay image under polar coordinates, transforming the polar coordinates into Cartesian coordinates, and finally obtain the birefringence image of the catheter PS-OCT sample.

[0023] The polarization leveling includes the following steps: disconnect the sample arm, first use the polarization controller 9 on the reference arm and the two polarization controllers in front of the polarization beam splitter to make the light intensity of the reference arm H path and V path equal in polarization concentration (the software has a monitoring function to make the light intensity of the H path and V path equal); secondly, connect the sample arm, use the polarization controller 3 in front of the polarization-maintaining optical fiber to use the catheter imaging as a reference, and adjust the peak values ​​of the front and rear surfaces of the dual-state imaging catheter on the one-dimensional image to be equal, so as to ensure that the light of the sample arm is evenly distributed to the two polarization controllers.

[0024] The background signal removal method is as follows: disconnect the sample arm, collect the A-scan information of the background noise and base frequency of the reference light, reconnect the sample arm and collect each A-scan of the signal, and subtract the background noise and base frequency of the reference light from each A-scan.

[0025] Beneficial effects of the present invention:

[0026] The conventional catheter optical coherence tomography system can only obtain the vascular intensity information, while the catheter polarization-sensitive optical coherence tomography system and demodulation method of the present invention can obtain the vascular birefringence information, thus providing more characteristic information for cardiovascular and cerebrovascular diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the structure of the system of the present invention.

[0028] Figure 2 It is a flow chart of the demodulation method of the present invention.

[0029] Figure 3 This is a catheter PS-OCT chicken breast tissue intensity image.

[0030] Figure 4 This is a birefringence image of chicken breast tissue acquired by catheter PS-OCT. DETAILED DESCRIPTION

[0031] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0032] Embodiment 1.

[0033] like Figure 1 shown.

[0034] A catheter polarization-sensitive optical coherence tomography system, such as Figure 1As shown, it includes a scanning light source 1, and the output light of the scanning light source 1 enters from port 1 of a 1:99 fiber coupler 2, and is distributed to the sample arm and the reference arm from ports 2 and 3 at a ratio of 1:99. The output light from port 2 of the 1:99 fiber coupler 2 enters the sample arm, and the light beam entering the sample arm is incident on the three-ring controller 3 and then incident on the polarization-maintaining fiber 4 with a length of 18.5 meters, and enters port 1 of the circulator 6. The light is emitted from port 2 of the circulator 6, and the output light enters the imaging catheter 11 through the rotating mechanism 8. The light reflected by the sample returns to the circulator 6 from the imaging catheter 11 and is emitted through port 3 of the circulator 6. The output 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 single-mode fiber 5 with a length of 18.5 meters. The output light enters port 1 of the circulator 7, and is emitted from port 2 to the reflective fiber delay line. The reflected light is incident through port 2 of the circulator 7 and is emitted from port 3 to the polarization controller 9. The outgoing light of the sample arm through port 3 of the circulator 6 and the outgoing light of the reference arm through the three-ring polarization controller 9 are respectively incident on the 50:50 fiber coupler 12 from ports 1 and 2 of the fiber coupler 12 to interfere, and enter the polarization controller 13 and the polarization controller 14 from ports 3 and 4 in a ratio of 50:50, and the outgoing light is respectively incident on polarization beam splitters 15 and 16, and the outgoing light of the polarization beam splitter 15 is respectively incident on balanced detectors 17 and 18 from ports 1 and 2, and the outgoing light of the polarization beam splitter 16 is respectively incident on balanced detectors 17 and 18 from ports 1 and 2, and the electrical signals of the balanced detectors 17 and 18 are received by the acquisition card 19 and transmitted to the computer 20.

[0035] The light source uses a fast scanning light source, and the system uses polarization-maintaining fiber to generate delays in orthogonal polarization states. Polarization diversity acquisition is performed through a polarization beam splitter. The length of the polarization-maintaining fiber depends on its birefringence to generate a phase delay equal to half the imaging depth of ordinary OCT. This method ensures that the system can simultaneously present polarization diversity imaging of two orthogonal input polarization states in one image, making it possible to subsequently eliminate the system birefringence changes introduced by catheter rotation.

[0036] Embodiment 2.

[0037] like Figure 2-4 shown.

[0038] A catheter polarization-sensitive optical coherence tomography demodulation method, such as Figure 2 As shown, it includes the following steps:

[0039] Step 1: Polarization leveling. On the one hand, equalize the light intensity of the reference light in the H path and the V path in the polarization concentration; on the other hand, evenly distribute the light of the sample arm to the two polarization states 13 and 14. The specific steps are as follows: disconnect the sample arm, first use the polarization controller 9 on the reference arm and the two polarization controllers in front of the polarization beam splitter to equalize the light intensity of the reference arm in the polarization concentration. The software monitors the light intensity of the H path and the V path. Secondly, connect the sample arm, use the polarization controller 3 in front of the polarization-maintaining fiber to use the catheter imaging as a reference, and adjust the peak values ​​of the front and rear surfaces of the dual-state imaging catheter on the one-dimensional image to be equal, so as to ensure that the light of the sample arm is evenly distributed to the two polarization states.

[0040] Step 2: Background signal removal. Background noise and fundamental frequency seriously affect the imaging quality, especially the fundamental frequency imaging. The background noise and fundamental frequency need to be removed. The specific method is: disconnect the sample arm (catheter part) and collect the A-scan information of the background noise and fundamental frequency of the reference light. After reconnecting the sample arm (catheter part), collect the signal for each A-scan, and subtract the background noise and fundamental frequency of the reference light from each A-scan.

[0041] Step 3: Two-state dispersion compensation, process the data collected by the PS-OCT system, perform Hilbert transform on each A-Line of the PS-OCT original image of channel H and channel V in polar coordinates, express the PS-OCT interference signal in the form of complex exponentials, and then perform numerical dispersion compensation on each A-Line after Hilbert transform, calculate the correction phase Φ(k) through the second-order and third-order dispersion coefficients, and multiply it with the original phase of the A-Line in the form of e exponentials, and realize numerical dispersion compensation of the original signal of the A-Line by phase addition and subtraction, and correct the frequency broadening of the interference signal. During dispersion compensation, the propagation constant β(Δk) is expanded at the central wave number k0 according to the Taylor series to obtain the second-order term coefficient and the third-order term coefficient. The second-order coefficient and the third-order coefficient of the two polarization states are respectively substituted into the power spectrum for order of magnitude and numerical traversal calculations to obtain the local optimal dispersion coefficient, and it is substituted into the phase compensation calculation formula of the interference signal to realize two-state dispersion compensation. The system is used to collect data in the no-load state of the catheter to obtain the interference signal of the reference light and the reflected light of the catheter probe end face; any A-Line data in the interference signal of the H channel or V channel is taken, and the data is Hilbert transformed to obtain the real part and imaginary part of the A-Line data, and then the A-Line signal is expressed in the form of amplitude index by Euler formula:

[0042]

[0043] Where φ0(k) is the initial phase of the interference signal;

[0044] The initial phase φ0(k) of the A-Line interference signal is expanded according to the Taylor series at the center wave number k0:

[0045] φ(k)=φ(k)-a2·(2πc) 2 (k-k0) 2 -a3·(2πc) 3 (k-k0) 3 … (2)

[0046] The second-order dispersion coefficient a2 and the third-order dispersion coefficient a3 are obtained, where a2 is the group velocity delay;

[0047] A set of a2 and a3 values ​​can be obtained for the phase of each input polarization state. The order of magnitude and value range of these two coefficients to be determined are initialized and substituted into formula (2) for traversal calculation. Within the initialized range, a set of compensation phases for interference terms are calculated. The calculated compensation phases are multiplied with the original interference signal index of the H channel or V channel in the form of e index. The obtained result is subjected to fast Fourier transformation, and the coordinate position of the maximum value of the power spectrum is taken as the order of magnitude or value of the dispersion coefficient. The outer surface area of ​​the fiber optic catheter probe wrapping layer is selected, and the order of magnitude and value of a2 and a3 are traversed in turn in the current area. The order of magnitude and value of a2 and a3 are determined by finding the maximum value of the interference peak of the interference signal during the traversal process. The obtained results a2 and a3 are the dual-state dispersion compensation coefficients of the current system. Since they are randomly selected A-Line data, a2 and a3 are actually the local optimal solutions of the dual-state dispersion compensation coefficients.

[0048] Step 4: Spectral shaping: To prevent the spectrum leakage of the range domain signal after Fourier transform, a cosine tapered window is used to shape the spectrum of the A-Line data.

[0049] Step 5: Perform interpolation Fourier transform. Due to the deep multiplexing in this system, the number of data points in the depth direction is small, so it is necessary to use interpolation Fourier transform to fill the number of points to make it consistent with the number of points on the circumference. Finally, the two-dimensional data information of PS-OCT is obtained, that is, the distance domain information H1+H2 and V1+V2 of the dual-state superposition of channel H and channel V.

[0050] Step 6: Two-state image segmentation, segment the distance domain information H1+H2 and V1+V2 of the two-state superposition of channel H and channel V, and perform cross-correlation operation on the two segments of each column of A-line data after selecting the segmentation point. Adjust the segmentation point by the cross-correlation peak frequency shift value, and finally obtain four groups of data H1, H2, V1 and V2. The two-state image segmentation method adopts the cross-correlation method to perform cross-correlation operation on the same A-Line data of the H channel and the V channel, which contains the A-Scan data of the two input polarization states. The starting position of the data of the two input polarization states is adjusted according to the offset of the cross-correlation peak to ensure that the first valid data point of the two input polarization states is aligned, and finally ensure that in the subsequent polarization solution, the four electric field data constituting the measured Jones matrix are from the same depth position of the same sample.

[0051] Step 7: Before performing polarization solution, select the reference position, that is, the position of the zero point of the birefringence phase delay of each column of A-line data. Generally, the sample surface is selected as the reference surface. However, in the catheter OCT system, the reflection intensity of the sample surface is unstable. Selecting an unstable reference surface will have a serious impact on the subsequent polarization information solution. In this system, the outer surface of the catheter sheath is selected as the reference surface, which is not affected by the sample to be tested. The specific selection method is to use the data frame to select the position of the outer surface of the catheter sheath, and the points with the largest intensity in the data frame for each column of A-line data are used as reference points.

[0052] Step 8: Substitute each column of A-line data of H1, H2, V1 and V2 into the polarization solution algorithm to obtain the birefringence phase delay.

[0053] Step 9: Perform coordinate interpolation transformation on the birefringence phase delay map in polar coordinates to convert polar coordinates into Cartesian coordinates, and finally obtain the birefringence image of the catheter PS-OCT sample

[0054] Compared with the unprocessed original intensity image, the output image processed by the present invention correctly expresses the birefringence information in the image. Figure 3 , 4 shown.

[0055] In specific implementation, the commonly used polarization information extraction method of PS-OCT is the Similar Jones Matrix (SJM) method. This method is based on the desktop PS-OCT and is improved according to the characteristics of catheter PS-OCT. However, the actual polarization solution result of this method is not ideal, and the birefringence information of biological tissues or samples cannot be correctly solved. For this reason, the present invention proposes a Similar Mueller Matrix (SMM) method. The core idea of ​​the SMM method is to convert the measured Jones matrix obtained by the catheter PS-OCT imaging system into a Mueller matrix, and then perform multiplication decomposition on the Mueller matrix to obtain a phase delay matrix. The phase delay of tissue birefringence is calculated using the trace of the phase delay matrix to achieve tissue birefringence imaging.

[0056] The calculation process is as follows:

[0057] The four electric field components of the system output are defined in Jones matrix form:

[0058]

[0059] The conversion relationship between Jones matrix and Mueller matrix is:

[0060]

[0061] in represents the Kronecker product, and U represents the transformation matrix:

[0062]

[0063] According to the conversion relationship between the Jones matrix and the Mueller matrix, the measured Jones matrix J output by the system is transformed into the measured Mueller matrix Q:

[0064]

[0065] The measured Mueller matrix Q Z By performing Euler transformation on the elements of , we can obtain:

[0066]

[0067] Take any single A-Scan and assume that the reference point on the A-Line is at a depth of z in the sample. The reference Mueller matrix Q obtained by measurement at z is Zref It can be expressed as:

[0068]

[0069] The data acquisition card measures the electrical signal at the polarization diversity point, which is output by the balanced photodetectors of the H channel and the V channel through photoelectric conversion. The collected electrical signals are subjected to background noise removal, two-state numerical dispersion compensation, spectrum shaping filtering and Fourier interpolation transformation to obtain two groups of signals H1+H2 and V1+V2 in the z domain of the H channel and the V channel. After image segmentation, the electric field signals H1, H2, V1, and V2 are obtained to form four polarization states. Taking a single A-Line as an example, the measurement Jones matrix J(z) obtained by constructing the imaging system at the sample depth z position is:

[0070]

[0071] Assume that the reference point on the A-Line is at a sample depth of z ref At the position, the corresponding reference Jones matrix J(z ref )for:

[0072]

[0073] Among them, for the catheter PS-OCT imaging system, when solving the polarization information of biological tissue, the reference surface can be the outer surface of the catheter or the front surface of the sample, and the measured Jones matrix J(z) and the reference Jones matrix J(z) can be used as the reference surface. ref ) is converted into the measurement Mueller matrix S(z) and the reference Mueller matrix S(z ref ).

[0074] Assume M ST is the round-trip transport matrix of the biological tissue or sample, M in and M out They represent the Mueller matrix of the system optical path, the reference Mueller matrix S(z ref ) is expressed as:

[0075]

[0076] The measured Mueller matrix S(z) at the sample z position measured at polarization diversity is expressed as:

[0077]

[0078] S(z ref ) and S(z) to eliminate the phase effect of the catheter on tissue birefringence when rotating at high speed. Formulas (2-14) and (2-15) can be rearranged to obtain the measured Mueller matrix M(z) of the biological tissue or sample at a depth of z. ref ,z):

[0079]

[0080] In order to construct M(z ref ,z) and M S,T is a pair of similar matrices, then it needs to satisfy Q Zref M out Must be an invertible matrix, but if Q Zref M out Including depolarization effect and bidirectional attenuation effect, the similarity matrix condition does not hold. In fact, refer to the Mueller matrix Q Zref is a unitary matrix, but in the catheter PS-OCT imaging system, the high-speed rotation of the optical fiber in the catheter will inevitably bring about a strong depolarization effect and a bidirectional attenuation effect, so M out The depolarization effect and bidirectional attenuation effect must also be included in the system optical path. It is necessary to use the matrix decomposition method to eliminate the depolarization effect and bidirectional attenuation effect in the system optical path.

[0081] Another important step in the SMM polarization solution method is to eliminate the depolarization effect and the bidirectional attenuation effect through matrix decomposition, and then obtain the measured phase delay Mueller matrix, calculate the phase delay of biological tissues or samples, and realize tissue birefringence imaging.

[0082] Common decomposition methods of the Mueller matrix include multiplication decomposition and summation decomposition. Multiplication decomposition is suitable for processing polarization units and systems of a priori sequences. Although the polarization units of biological tissues are discrete, complex and diverse, and obviously do not have the characteristics of a priori sequences, the decomposed Mueller matrix obtained by using the LC Mueller matrix decomposition method can basically correctly describe the basic polarization characteristics of biological tissues. Based on the catheter PS-OCT imaging system, the Mueller matrix can also be decomposed using the symmetric decomposition method. When it is known that the Mueller has been depolarized, the symmetric decomposition method can also accurately express the basic polarization characteristics of biological tissues. This method uses the LC Mueller matrix decomposition method to obtain the phase delay Mueller matrix M of the measured Mueller matrix R .

[0083] Any Mueller matrix describing a light-transmitting medium includes the three polarization effects of the target light-transmitting medium modulating the incident light, including bidirectional attenuation, phase delay or depolarization. Any Mueller matrix can be decomposed into the following form:

[0084] M=M Δ M R M D (2-17)

[0085] Where M is an arbitrary Mueller matrix, M Δ It represents the depolarization matrix of the incident light modulated by the target light-transmitting medium, which describes the ability of the target light-transmitting medium to depolarize the incident light. RIt represents the phase delay matrix of the incident light modulated by the target light transmission medium. It describes that after the incident light passes through the polarization device, the two intrinsic polarization states generated have different propagation speeds in the fast and slow axes, resulting in a time delay in the time domain signals of the two intrinsic polarization states. The conversion to the frequency domain is the phase delay. D It represents the two-way attenuation matrix of the incident light modulated by the target light-transmitting medium, and describes the relationship between the reflected light intensity of the incident light passing through the light-transmitting medium and the polarization state of the incident light.

[0086] Phase delay matrix M R It can be expressed as:

[0087]

[0088] Where m R is the phase delay matrix M R A 3×3 submatrix of , which can be calculated by formula (2-19):

[0089]

[0090] where m' and m Δ They are the depolarization matrices M Δ The product of the depolarization matrix and the phase delay matrix is ​​M', which can be expressed as:

[0091]

[0092] In the unknown depolarization matrix M Δ and the phase delay matrix M R In the case of D The two-way attenuation matrix M is calculated by multiplying the measured Mueller matrix M D It can be expressed as:

[0093]

[0094] in, represents the two-way attenuation vector, m D is the two-way attenuation matrix M D A 3×3 sub-matrix of . The two-way attenuation vector It can be defined as:

[0095]

[0096] in is the magnitude of the bidirectional attenuation vector, Its unit vector, d i Represents the components of the two-way attenuation vector, which can be calculated by formula (2-23). ​​In the context formula, M 1(i+1)(i=1, 2, 3) represents the elements of the first row of the measurement Mueller matrix M.

[0097]

[0098] Submatrix m of the two-way decay matrix D It can be expressed as:

[0099]

[0100] in ||·|| represents the Euclidean vector norm. From the above, the m' matrix can be calculated.

[0101] Depolarization matrix M Δ It can be expressed as:

[0102]

[0103] in Represents the polarization vector of the depolarization matrix, whose magnitude depends on the polarization vector and the two-way attenuation vector

[0104]

[0105]

[0106] In the phase delay matrix M R In the calculation process, it is not necessary to solve the debiasing matrix M Δ , we only need to solve its submatrix m Δ That is, the submatrix m of the debiased matrix Δ It can be calculated by formula (2-28):

[0107]

[0108] where λ i (i=1, 2, 3) represents the eigenvalues ​​of the submatrix m'.

[0109] At this point, the submatrix m of the debiasing matrix is ​​obtained Δ , m Δ Substituting m' into formula (2-19), the phase delay matrix M can be solved by combining formula (2-18) R In the catheter PS-OCT imaging system, the Mueller matrix M(z ref ,z) to perform matrix multiplication decomposition to obtain the Mueller matrix containing only tissue birefringence, which has eliminated the depolarization effect and the bidirectional attenuation effect, and the phase delay matrix M R (z ref ,z) is:

[0110]

[0111] The measured phase delay matrix M R (z ref ,z), biological tissue or sample transmission matrix and system transmission output matrix are Mueller matrices that only contain birefringence components, and at this time is a unitary matrix, satisfying the similarity matrix condition, M R (z ref ,z) and That is, a pair of similar matrices. The total phase delay and the decomposed phase delay matrix M R The relationship can be expressed as:

[0112]

[0113] in, It means the transmission Mueller matrix of the biological tissue or sample at the depth z containing only the birefringence component, tr means Trace of the matrix.

[0114] In specific implementation, spectrum shaping can be performed using the following methods: time-frequency localization is achieved by signal windowing. Signal windowing can correct the problem of spectrum leakage. Selecting different window functions will ultimately affect the resolution and signal-to-noise ratio. Let the k-domain interference signal pass through a cosine tapered window for signal windowing. The windowed signal is expressed as:

[0115]

[0116]

[0117] Where y represents the output sequence, that is, the windowed x sequence, n is the number of elements in the x sequence, r is the ratio of the total length of the tapered part to the total length of the signal, and m is rounded down; when r≤0, the window is equivalent to a rectangular window; when r≥1, the window is equivalent to a Hanning window. By setting the target resolution requirement, n and r can be traversed according to the above process.

[0118] The parts not involved in the present invention are the same as the prior art or can be implemented by using the prior art.

Claims

1. A catheter polarization-sensitive optical coherence tomography demodulation method, characterized in that it The following steps are involved: Step 1: polarization leveling, on the one hand, equalizing the light intensity of the reference light in the H path and the V path in the polarization concentration; on the other hand, evenly distributing the light of the sample arm to the two polarization controllers (13, 14); Step 2: Background signal removal. Background noise and baseband seriously affect the imaging quality, especially baseband imaging. The background noise and baseband need to be removed. Step 3: Dual-state dispersion compensation, process the data collected by the PS-OCT system, perform Hilbert transform on each A-Line of the PS-OCT original image of channel H and channel V in polar coordinates, express the PS-OCT interference signal in the form of complex exponentials, and then perform numerical dispersion compensation on each A-Line after Hilbert transform, calculate the correction phase Φ(k) through the second-order and third-order dispersion coefficients, and multiply it with the original phase of the A-Line in the form of e-exponential, and realize numerical dispersion compensation of the original signal of the A-Line by phase addition and subtraction, and correct the frequency broadening of the interference signal; During dispersion compensation, the propagation constant β(Δk) is expanded at the central wave number k0 according to the Taylor series to obtain the second-order term coefficient and the third-order term coefficient. The second-order coefficient and the third-order coefficient of the two polarization states are respectively substituted into the power spectrum for order of magnitude and numerical traversal calculation to obtain the local optimal dispersion coefficient, which is then substituted into the phase compensation calculation formula of the interference signal to achieve dual-state dispersion compensation. The system is used to collect data in the catheter no-load state to obtain the interference signal of the reference light and the reflected light of the catheter probe end face. Any A-Line data in the interference signal of the H channel or V channel is taken, and the data is Hilbert transformed to obtain the real and imaginary parts of the A-Line data, and then the A-Line signal is expressed in the form of amplitude index by the Euler formula: Where φ0(k) is the initial phase of the interference signal; The initial phase φ0(k) of the A-Line interference signal is expanded according to the Taylor series at the center wave number k0: φ(k)=φ(k)-a2·(2πc) 2 ·(k-k0) 2 -a3·(2πc) 3 ·(k-k0) 3 … (2) The second-order dispersion coefficient a2 and the third-order dispersion coefficient a3 are obtained, where a2 is the group velocity delay; A set of a2 and a3 values ​​can be obtained for the phase of each input polarization state. The order of magnitude range and the numerical range of these two unknown coefficients are initialized and substituted into formula (2) for traversal calculation. Within the initialized range, a set of compensation phases for interference terms are calculated. The calculated compensation phases are multiplied with the original interference signal index of the H channel or V channel in the form of e index. The obtained result is subjected to fast Fourier transformation, and the coordinate position of the maximum value of the power spectrum is taken as the order of magnitude or value of the dispersion coefficient. The outer surface area of ​​the fiber optic catheter probe wrapping layer is selected, and the order of magnitude and numerical size of a2 and a3 are traversed in the current area in turn. The order of magnitude and numerical size of a2 and a3 are determined by finding the maximum value of the interference peak of the interference signal during the traversal process. The obtained results a2 and a3 are the dual-state dispersion compensation coefficients of the current system. Since they are randomly selected A-Line data, a2 and a3 are actually the local optimal solutions of the dual-state dispersion compensation coefficients. Step 4: Spectral shaping: After Fourier transform, the spectrum of the distance domain signal is not leaked. For A-Line data, a cosine tapered window is used to shape the spectrum. Step 5: Perform interpolation Fourier transform. Due to the deep multiplexing in this system, the number of data points in the depth direction is small, and the interpolation Fourier transform is needed to fill the number of points to make it consistent with the number of points on the circumference, and finally obtain the two-dimensional data information of PS-OCT, that is, the distance domain information H1+H2 and V1+V2 of the dual-state superposition of channel H and channel V; Step 6: Two-state image segmentation, segment the distance domain information H1+H2 and V1+V2 of the two-state superposition of channel H and channel V, and perform cross-correlation operation on the two segments of each column of A-line data after selecting the segmentation point, and adjust the segmentation point by the cross-correlation peak frequency shift value, and finally obtain four groups of data H1, H2, V1 and V2; the two-state image segmentation method adopts the cross-correlation method, and performs cross-correlation operation on the same A-Line data of the H channel and the V channel, which contains A-Scan data of two input polarization states. The starting position of the data of the two input polarization states is adjusted according to the offset of the cross-correlation peak to ensure that the first valid data points of the two input polarization states are aligned, and finally ensure that in the subsequent polarization solution, the four electric field data constituting the measurement Jones matrix are all from the same depth position of the same sample; Step 7: Before performing polarization solution, select the reference position, that is, the position of the zero point of the birefringence phase delay of each column of A-line data. Generally, the sample surface is selected as the reference surface. However, in the catheter OCT system, the reflection intensity of the sample surface is unstable. Selecting an unstable reference surface will have a serious impact on the subsequent polarization information solution. Select the outer surface of the catheter sheath as the reference surface, which is not affected by the sample to be tested. Use the data frame to select the position of the outer surface of the catheter sheath. The points with the largest intensity in each column of A-line data in the data frame are used as reference points. Step 8: Substitute each column of A-line data of H1, H2, V1 and V2 into the polarization solution algorithm to obtain the birefringence phase delay; Step nine: performing coordinate interpolation transformation on the birefringence phase delay image under polar coordinates to transform the polar coordinates into Cartesian coordinates, and finally obtaining the birefringence image of the catheter PS-OCT sample; Compared with the unprocessed original intensity image, the birefringence information in the processed output image is correctly expressed; The polarization information extraction method used by PS-OCT is the Similar Jones Matrix (SJM) method. This method is based on the desktop PS-OCT and is improved according to the characteristics of catheter PS-OCT. The similar Mueller Matrix (SMM) method is used. The core idea of ​​the SMM method is to convert the measured Jones matrix obtained by the catheter PS-OCT imaging system into a Mueller matrix, and then perform multiplication decomposition on the Mueller matrix to obtain a phase delay matrix. The phase delay of tissue birefringence is calculated using the trace of the phase delay matrix to achieve tissue birefringence imaging. The calculation process is as follows: The four electric field components of the system output are defined in Jones matrix form: The conversion relationship between Jones matrix and Mueller matrix is: in represents the Kronecker product, and U represents the transformation matrix: According to the conversion relationship between the Jones matrix and the Mueller matrix, the measured Jones matrix J output by the system is transformed into the measured Mueller matrix Q: The measured Mueller matrix Q Z By performing Euler transformation on the elements of , we can obtain: Take any single A-Scan and assume that the reference point on the A-Line is at a depth of z in the sample. The reference Mueller matrix Q obtained by measurement at z is Zref It can be expressed as: The data acquisition card measures the electrical signal at the polarization diversity point, which is output by the balanced photodetectors of the H channel and the V channel through photoelectric conversion. The collected electrical signals are subjected to background noise removal, two-state numerical dispersion compensation, spectrum shaping filtering and Fourier interpolation transformation to obtain two groups of signals H1+H2 and V1+V2 in the z domain of the H channel and the V channel. After image segmentation, the electric field signals H1, H2, V1, and V2 are obtained to form four polarization states. Taking a single A-Line as an example, the measurement Jones matrix J(z) obtained by constructing the imaging system at the sample depth z position is: Assume that the reference point on the A-Line is at a sample depth of z ref At the position, the corresponding reference Jones matrix J(z ref )for: Among them, for the catheter PS-OCT imaging system, when solving the polarization information of biological tissue, the reference surface can be the outer surface of the catheter or the front surface of the sample, and the measured Jones matrix J(z) and the reference Jones matrix J(z) can be used as the reference surface. ref ) is converted into the measurement Mueller matrix S(z) and the reference Mueller matrix S(z ref ); Assume M ST is the round-trip transport matrix of the biological tissue or sample, M in and M out They represent the Mueller matrix of the system optical path, the reference Mueller matrix S(z ref ) is expressed as: The measured Mueller matrix S(z) at the sample z position measured at polarization diversity is expressed as: S(z ref ) and S(z) to eliminate the phase effect of the catheter on tissue birefringence when rotating at high speed. Formulas (2-14) and (2-15) can be rearranged to obtain the measured Mueller matrix M(z) of the biological tissue or sample at a depth of z. ref ,z): In order to construct M(z ref ,z) and M S,T is a pair of similar matrices, then it needs to satisfy Q Zref M out Must be an invertible matrix, but if Q Zref M out Including depolarization effect and bidirectional attenuation effect, the similarity matrix condition does not hold; in fact, referring to the Mueller matrix Q Zref is a unitary matrix, but in the catheter PS-OCT imaging system, the high-speed rotation of the optical fiber in the catheter will inevitably bring about a strong depolarization effect and a bidirectional attenuation effect, so M out The depolarization effect and the bidirectional attenuation effect must also be included in the system, so it is necessary to eliminate the depolarization effect and the bidirectional attenuation effect in the system optical path through the matrix decomposition method; Another important step in the SMM polarization solution method is to eliminate the depolarization effect and the bidirectional attenuation effect through matrix decomposition, and then obtain the measured phase delay Mueller matrix, calculate the phase delay of biological tissues or samples, and realize tissue birefringence imaging; Common decomposition methods of the Mueller matrix include multiplication decomposition and summation decomposition. Multiplication decomposition is suitable for processing polarization units and systems of a priori sequences. Although the polarization units of biological tissues are discrete, complex and diverse, and obviously do not have the characteristics of a priori sequences, the decomposed Mueller matrix obtained by using the LC Mueller matrix decomposition method can basically correctly describe the basic polarization characteristics of biological tissues. Based on the catheter PS-OCT imaging system, the Mueller matrix is ​​decomposed using the symmetric decomposition method. When it is known that the Mueller has been depolarized, the symmetric decomposition method can also accurately express the basic polarization characteristics of biological tissues. The phase delay Mueller matrix M of the measured Mueller matrix is ​​obtained by using the LC Mueller matrix decomposition method. R ; Any Mueller matrix describing a light-transmitting medium includes the three polarization effects of the target light-transmitting medium modulating the incident light, including bidirectional attenuation, phase delay or depolarization; any Mueller matrix can be decomposed into the following form: M=M Δ M R M D (2-17) Where M is an arbitrary Mueller matrix, M Δ M represents the depolarization matrix of the incident light modulated by the target light-transmitting medium, which describes the ability of the target light-transmitting medium to depolarize the incident light; R It represents the phase delay matrix of the incident light modulated by the target light transmission medium. It describes that after the incident light passes through the polarization device, the two intrinsic polarization states generated have different propagation speeds in the fast and slow axes, resulting in a time delay in the time domain signals of the two intrinsic polarization states. The conversion to the frequency domain is the phase delay. D It represents the two-way attenuation matrix of the incident light modulated by the target light transmission medium, which describes the relationship between the reflected light intensity of the incident light passing through the light transmission medium and the polarization state of the incident light; Phase delay matrix M R It can be expressed as: Where m R is the phase delay matrix M R A 3×3 submatrix of , which can be calculated by formula (2-19): where m' and m Δ They are the depolarization matrices M Δ The product of the depolarization matrix and the phase delay matrix is ​​M', which can be expressed as: In the unknown depolarization matrix M Δ and the phase delay matrix M R In the case of D The two-way attenuation matrix M is calculated by multiplying the measured Mueller matrix M D It can be expressed as: in, represents the two-way attenuation vector, m D is the two-way attenuation matrix M D A 3×3 submatrix of ; a two-way attenuation vector It can be defined as: in is the magnitude of the bidirectional attenuation vector, Its unit vector, d i Represents the components of the two-way attenuation vector, which can be calculated by formula (2-23). ​​In the context formula, M 1(i+1) (i=1, 2, 3) represents the elements of the first row of the measurement Mueller matrix M; Submatrix m of the two-way attenuation matrix D It can be expressed as: in ‖·‖ represents the Euclidean vector norm; above, the m' matrix can be calculated; Depolarization matrix M Δ It can be expressed as: in Represents the polarization vector of the depolarization matrix, whose magnitude depends on the polarization vector and the two-way attenuation vector In the phase delay matrix M R In the calculation process, it is not necessary to solve the debiasing matrix M Δ , we only need to solve its submatrix m Δ That is, the submatrix m of the debiased matrix Δ It can be calculated by formula (2-28): where λ i (i=1,2,3) represents the eigenvalue of submatrix m'; At this point, the submatrix m of the debiasing matrix is ​​obtained Δ , m Δ Substituting m' into formula (2-19), the phase delay matrix M can be solved by combining formula (2-18) R In the catheter PS-OCT imaging system, the Mueller matrix M(z ref ,z) to perform matrix multiplication decomposition to obtain the Mueller matrix containing only tissue birefringence, which has eliminated the depolarization effect and the bidirectional attenuation effect, and the phase delay matrix M R (z ref ,z) is: The measured phase delay matrix M R (z ref ,z), biological tissue or sample transmission matrix and system transmission output matrix are Mueller matrices that only contain birefringence components, and at this time is a unitary matrix, satisfying the similarity matrix condition, M R (z ref ,z) and That is, a pair of similar matrices; the total phase delay and the decomposed phase delay matrix M R The relationship can be expressed as: in, It means the transmission Mueller matrix of the biological tissue or sample at the depth z containing only the birefringence component, tr means Trace of a matrix; Spectral shaping can be performed by the following methods: time-frequency localization is achieved by signal windowing. Signal windowing can correct the problem of spectrum leakage. Selecting different window functions will eventually affect the resolution and signal-to-noise ratio. The k-domain interference signal is windowed through a cosine tapered window. The windowed signal is expressed as: Where y represents the output sequence, that is, the windowed x sequence, n is the number of elements in the x sequence, r is the ratio of the total length of the tapered part to the total length of the signal, and m is rounded down; when r≤0, the window is equivalent to a rectangular window; when r≥1, the window is equivalent to a Hanning window. By setting the target resolution requirement, n and r can be traversed according to the above process.

2. The method according to claim 1, characterized in that The polarization leveling comprises the following steps: disconnecting the sample arm, firstly using the polarization controller (9) on the reference arm and the two polarization controllers in front of the polarization beam splitter to make the light intensities of the reference arm H path and V path equal in polarization concentration; secondly, connecting the sample arm, using the polarization controller (3) in front of the polarization-maintaining optical fiber to take the catheter imaging as a reference, and equalizing the peak values ​​of the front and rear surfaces of the dual-state imaging catheter on the one-dimensional image, thereby ensuring that the light of the sample arm is evenly distributed to the two polarization controllers.

3. The method according to claim 1, characterized in that The background signal removal method is as follows: disconnect the sample arm, collect the A-scan information of the background noise and base frequency of the reference light, reconnect the sample arm and collect each A-scan of the signal, and subtract the background noise and base frequency of the reference light from each A-scan.

Images