Analysis method of vascular wall shear stress based on intravascular imaging and imaging device

Abstract

The invention discloses a blood vessel wall shear stress analysis method based on intravascular imaging and an imaging device. The method comprises the following steps: acquiring IVOCT spectral signals in two groups of target blood vessels; removing blood vessel walls and artifacts in the second group of spectral signals by utilizing a three-dimensional blood vessel morphological structure established by the first group of spectral signals to obtain an intracavity depth domain complex signal; doppler frequency shift calculation is carried out on the intracavitary depth domain complex signal, phase unwrapping is carried out, and an intravascular flow velocity distribution image is obtained; and calculating the wall shear stress of the blood vessel by using the intravascular flow velocity distribution image to obtain a WSS image of the wall shear stress of the blood vessel. According to the invention, detection of intravascular structures and intravascular blood flow velocity distribution can be realized, and analysis of vascular wall shear stress can be realized. The method has the advantages of being high in imaging speed, high in spatial resolution, high in speed and shear force sensitivity, low in calculation cost and the like, multi-dimensional detection and analysis of vasculopathy can be achieved, and the method has important significance in predicting occurrence and development of diseases clinically.

Description

Technical Field

[0001] The present invention relates to the technical field of vascular imaging, and in particular to an analysis method and an imaging device for vascular wall shear stress based on intravascular imaging. Background Art

[0002] Atherosclerosis or coronary artery disease (CAD) is the most common cardiovascular disease and remains the leading cause of death worldwide. Atherosclerotic lesions are usually manifested as plaque accumulation in the vascular membrane. Interventional imaging technology provides the function of visualizing the microstructure of plaques, which helps to further explore the pathogenesis of atherosclerotic lesions and provides a new solution for percutaneous coronary intervention (PCI) in clinical practice.

[0003] Intravascular Optical Coherence Tomography (IVOCT) has been widely used in cardiovascular imaging and can provide high-resolution (1-15um) cross-sectional images of the vascular wall to accurately analyze tissue structure and identify plaque components. Although the current IVOCT technology has significant advantages in intravascular detection, new imaging methods still need to be explored. By obtaining functional intravascular blood flow velocity distribution information combined with structural information, multi-dimensional information on vascular lesions and complications can be provided, such as identifying stenosis caused by atherosclerotic plaques, stratifying atherosclerotic plaque risk, assessing the risk of CAD patients, and monitoring treatment before and after stent implantation.

[0004] Compared with traditional intravascular optical coherence tomography technology, the calculation and analysis methods of intravascular blood flow distribution have always been a major bottleneck restricting the development of intravascular optical coherence tomography technology due to the influence of image noise.

[0005] In addition, high vascular wall shear stress (WSS) is an important factor causing plaque rupture, and low WSS is an important factor causing atherosclerosis. Through WSS calculation, information such as wall shear stress gradient (WSSG) and oscillation shear index (OSI) can also be obtained, which can provide multi-dimensional data basis for the analysis and prediction of intravascular diseases. The various parameters of vascular WSS can be obtained by measuring the intravascular flow velocity distribution and then calculating the OCT Doppler blood flow distribution gradient field through computational fluid dynamics (CFD), which can be used as a predictor of plaque rupture in patients with high risk of coronary heart disease to accurately guide PCI.

[0006] However, the existing WSS measurement technology needs to rely on the fusion of IVOCT / intravascular ultrasound (IVUS) and angiography technology and perform CFD. This method requires precise registration of the two modal images and has high computational costs. Secondly, WSS can also be solved by magnetic resonance imaging (MRI), which has low resolution and cannot obtain intravascular images.

[0007] Existing technologies limit the development of WSS analysis in the cardiovascular field. Currently, there is an urgent need to complete the calculation of intravascular blood flow velocity distribution and WSS analysis through single-modality imaging technology. Summary of the invention

[0008] Therefore, the object of the present invention is to provide a method and an imaging device for analyzing vascular wall shear stress based on intravascular imaging, which can calculate vascular wall shear stress using a single IVOCT mode.

[0009] In order to achieve the above object, a method for analyzing vascular wall shear stress based on intravascular imaging of the present invention comprises the following steps:

[0010] S1, collecting IVOCT spectral signals in two groups of target blood vessels;

[0011] S2, using the three-dimensional vascular morphology structure established by the first group of spectral signals, removing the vascular wall and artifacts in the second group of spectral signals, and obtaining a complex signal in the cavity depth domain;

[0012] S3, performing Doppler frequency shift calculation on the intracavitary depth domain complex signal and performing phase unwrapping to obtain an intravascular flow velocity distribution image;

[0013] S4. Calculate the vascular wall shear stress using the intravascular flow velocity distribution image to obtain the vascular wall shear stress WSS image.

[0014] Further preferably, in S1, collecting IVOCT spectral signals in two groups of target blood vessels includes:

[0015] Process 1: Place the imaging catheter at a predetermined position in the target blood vessel, flush an appropriate amount of saline or contrast agent into the blood vessel, and the catheter starts to rotate. When there is no blood obstruction in real-time imaging, the catheter is withdrawn to obtain the first spectral signal during the withdrawal process;

[0016] Process 2: The imaging catheter is placed again at a predetermined position in the target blood vessel, the catheter starts to rotate, an IVOCT spectral image is acquired, an electrocardiogram signal is acquired synchronously, and all the acquired data within one heartbeat cycle is acquired as a second spectral signal.

[0017] Further preferably, in S2, the three-dimensional vascular morphology structure established by the first group of spectral signals is used to eliminate the vascular wall and artifacts in the second group of spectral signals to obtain the intracavitary depth domain complex signal, including the following steps:

[0018] S201, performing Fourier transform on the second spectral image to obtain a depth domain complex signal of IVOCT;

[0019] S202, extracting intensity components from the depth domain complex signal based on IVOCT;

[0020] S203, using intensity components to divide the structural information of the vascular tissue to obtain a structural image, aligning the structural image with the three-dimensional vascular morphological structure, and removing the vascular wall and artifacts in the structural image using the three-dimensional vascular morphological structure to obtain an intracavitary structural image and an intracavitary depth domain complex signal.

[0021] Further preferably, after S203, the method further includes: subtracting the phase information of the inner and outer walls of the sheath or the self-coherent signal inside the system from the phase component extracted based on the intracavitary depth domain complex signal to obtain a corrected phase component.

[0022] Further preferably, in S3, Doppler frequency shift calculation is performed on the intracavity depth domain complex signal, and phase unwrapping is performed, including:

[0023] S301, using the phase component of the complex signal in the cavity depth domain after correction, calculate the Doppler frequency shift to obtain a wrapped phase map;

[0024] S302, calculating a reliability index according to the second-order derivative of the wrapped phase of each pixel in the wrapped phase image;

[0025] S303, gradually downsampling the wrapped phase image to generate phase images with different resolutions;

[0026] S304, starting from the point with the highest reliability in the phase map, unpacking, generating a priority queue for adjacent pixel points in descending order of reliability, and then selecting the point with the highest reliability in the priority queue for unpacking each time, and updating the path;

[0027] S305, for each point on the path, correct the jump according to the unwrapped phase of the adjacent pixel points;

[0028] S306, extracting global phase features from the coarse-resolution layer of the unwrapped phase image; refining path selection in the medium-resolution layer, and accurately adjusting local details in the high-resolution layer;

[0029] S307 , smoothing the unwrapped phase image to obtain an unwrapped phase image.

[0030] Further preferably, in S302, the reliability index is calculated according to the following formula:

[0031] Reliability Index

[0032]

[0033] Among them, H x (i, j) is the horizontal second-order derivative of pixel (i, j), H y (i, j) is the second-order derivative in the vertical direction of pixel (i, j), and ∈ is a positive number less than 10 to avoid the denominator being zero.

[0034] Further preferably, in S3, the phase difference between any two adjacent A-lines is calculated from the unwrapped phase signal. A phase image is obtained, and the intravascular flow velocity is calculated according to the phase image according to the following formula:

[0035]

[0036] Where Δf (i,j) is the flow velocity, λ is the central wavelength of the light source, n is the refractive index of blood in the tissue, τ is the time interval between adjacent A-lines, θ is the Doppler angle when measuring the flow velocity, which is determined by the angle between the focusing beam direction of the catheter and the blood flow direction, i and j are the horizontal and vertical coordinates in the polar coordinate image respectively; φ unwrapped (i, j) is the phase difference between two adjacent A-lines.

[0037] Further preferably, in S4, the blood vessel wall shear stress is calculated using the blood vessel flow velocity distribution image, including: according to the fact that the shear stress is proportional to the flow velocity gradient in the fluid, the shear stress is calculated according to the following formula:

[0038]

[0039] Where du / dx and du / dy are the derivatives of the blood flow velocity components in the x and y directions in polar coordinates.

[0040] The present invention also provides an imaging device for analyzing vascular wall shear stress based on intravascular imaging, which is used to implement the steps of the above-mentioned vascular wall shear stress analysis method based on intravascular imaging, including: a swept-frequency light source, a reference arm, an imaging device and an image processing module; a part of the swept-frequency light emitted by the swept-frequency light source enters the reference arm to form a standard signal, and the other part of the light enters the imaging device and irradiates the sample to form a sample signal; the sample signal and the standard signal are interfered to form an IVOCT spectral image; the spectral image includes a first spectral signal collected under the condition of using a contrast agent and a second spectral signal collected without using a contrast agent; the three-dimensional vascular morphological structure established by the collected first spectral signal is eliminated by using the image processing module, and the vascular wall and artifacts in the second spectral signal are obtained to obtain an intracavitary depth domain complex signal; the intracavitary depth domain complex signal is phase unwrapped to obtain an intravascular flow velocity distribution image; the intravascular flow velocity distribution image is used to calculate the vascular wall shear stress to obtain a vascular wall shear stress WSS image.

[0041] The analysis method and imaging device of vascular wall shear stress based on intravascular imaging disclosed in the present application can realize the detection of intravascular structure and intravascular blood flow velocity distribution, and can also realize the analysis of vascular wall shear stress. It has the characteristics of fast imaging speed, high spatial resolution, high sensitivity to velocity and shear force, low calculation cost, etc., and can realize multi-dimensional detection and analysis of vascular lesions, which is of great significance in clinical prediction of the occurrence and development of diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of a method for analyzing vascular wall shear stress based on intravascular imaging of the present invention;

[0043] Figure 2 This is a schematic diagram of the process of collecting IVOCT spectral images in the target blood vessel in the present invention;

[0044] Figure 3 It is a schematic diagram of an imaging device for analyzing the shear stress of the blood vessel wall based on intravascular imaging of the present invention;

[0045] Figure 4 It is a phase unwrapping flow chart of the present invention;

[0046] Figure 5 (a) is a structural image of a certain cross section of a blood vessel phantom according to an embodiment of the present invention;

[0047] Figure 5(b) is a flow velocity distribution image before phase correction in the blood vessel cavity of the cross section in an embodiment of the present invention;

[0048] Figure 5 (c) is the velocity distribution image after phase correction in the blood vessel cavity of the cross section of the present invention;

[0049] Figure 5 (d) is the velocity distribution image after phase unwrapping in the blood vessel cavity of the cross section of the present invention;

[0050] Figure 6 It is a graph of the local average flow velocity of an exemplary embodiment of the present invention.

[0051] Wherein: 201-light source; 202-1×2 fiber coupler; 203-first fiber circulator; 204-reference arm fiber collimator; 205-objective lens; 206-flat high reflector; 207-polarization controller; 208-2×2 fiber coupler; 209-photoelectric balance detector; 210-signal processing unit; 211-second fiber circulator; 212-fiber rotation and retraction unit; 213-first torque spring; 214-first focusing element; 215-first reflecting element; 216-first sample; 217-retraction unit; 218-second torque spring; 219-second focusing element; 220-second reflecting element; 221-remote micromotor; 222-second sample. DETAILED DESCRIPTION

[0052] The present invention is further described in detail below through the accompanying drawings and specific embodiments.

[0053] like Figure 1 As shown, an analysis method of vascular wall shear stress based on intravascular imaging is provided in an embodiment of one aspect of the present invention, comprising the following steps:

[0054] S1. Collect two groups of IVOCT spectral images in the target blood vessels, wherein the first group of spectral images is collected under the condition of using a contrast agent, and the collected data is a first spectral signal, and the second group of spectral images is collected under the condition of not using a contrast agent, and the collected data is a second spectral signal;

[0055] S2, using the three-dimensional vascular morphology structure established by the first spectral signal to eliminate the vascular wall and artifacts in the second spectral signal to obtain a complex signal in the cavity depth domain;

[0056] S3, performing Doppler frequency shift calculation on the intracavitary depth domain complex signal and performing phase unwrapping processing to obtain an intravascular flow velocity distribution image;

[0057] S4. Calculate the vascular wall shear stress using the intravascular flow velocity distribution image to obtain the vascular wall shear stress WSS image.

[0058] like Figure 2 As shown, in S1, the acquisition of the IVOCT spectral image in the target blood vessel includes two acquisition processes:

[0059] Process 1:

[0060] D01, placing the imaging catheter at a predetermined position in the target blood vessel, D02, flushing an appropriate amount of physiological saline or contrast agent into the blood vessel, and the catheter starts to rotate, D03, when there is no blood obstruction in real-time imaging, the catheter is withdrawn, and the first spectral signal during the withdrawal process is obtained;

[0061] Process 2:

[0062] D04, placing the imaging catheter again at a predetermined position in the target blood vessel;

[0063] D05. The catheter starts to rotate, and the IVOCT spectral image is acquired. The electrocardiogram signal is acquired synchronously, and all the collected data within one heartbeat cycle is acquired as the second spectral signal. It should be noted that the second spectral signal can be collected either by retracting or in situ. For the proximal drive catheter, the data during the heartbeat is eliminated; for the distal drive catheter, it is retracted through ultra-high-speed rotation, and all data is collected within one heartbeat cycle. Since the proximal drive scanning speed is slow, it is easily affected by the beating of the heart. The inter-frame correlation based on the retracted image is used to select the appropriate threshold to filter out the data to be eliminated.

[0064] It should be noted that no contrast agent is injected during the second withdrawal. The purpose is to remove the image of the injected contrast agent and obtain the true blood flow velocity.

[0065] In S2, the 3D vascular morphology structure established by the first spectral signal is used to remove the vascular wall and artifacts in the second spectral signal to obtain an intracavitary complex domain signal, including the following steps:

[0066] S201, establishing a three-dimensional blood vessel morphological structure based on the first spectral signal;

[0067] S202, performing Fourier transform on the second spectral signal to obtain a depth domain complex signal of IVOCT;

[0068] S203, extracting intensity components from the depth domain complex signal based on IVOCT;

[0069] S204, using the intensity component to divide the structural information of the vascular tissue to obtain a structural image, aligning the structural image with the three-dimensional vascular morphological structure, and removing the vascular wall and artifacts in the structural image using the three-dimensional vascular morphological structure to obtain an intracavitary structural image, and correspondingly, obtaining an intracavitary depth domain complex signal;

[0070] S205, phase correction, extracting the phase component based on the intracavitary depth domain complex signal, subtracting the phase information of the inner and outer walls of the sheath or the self-coherent signal inside the system, to obtain a corrected phase component.

[0071] Phase correction can make the phase difference signal between adjacent A lines more uniform, thereby achieving the effect of suppressing catheter motion artifacts.

[0072] The vascular walls and artifacts in the structural image can be eliminated based on a phase artifact correction algorithm that tracks the signal of the catheter sheath wall or the reflection surface of the optical element. First, the position index of the catheter sheath wall or the reflection surface of the optical element is found through a boundary recognition algorithm, and then the phase of each pixel in the image is subtracted from the phase of the catheter sheath wall or the reflection surface of the optical element to correct the phase disturbance caused by the non-uniform distortion of the catheter.

[0073] like Figure 4 As shown, in S3, the Doppler frequency shift calculation is performed on the intracavity depth domain complex signal, and phase unwrapping processing is performed, including:

[0074] S301, using the phase component of the complex signal in the cavity depth domain after correction, calculate the Doppler frequency shift to obtain a wrapped phase map;

[0075] S302, calculating a reliability index according to the second-order derivative of the wrapped phase of each pixel in the wrapped phase image;

[0076] S303, gradually downsampling the wrapped phase image to generate phase images with different resolutions;

[0077] S304, starting from the point with the highest reliability in the phase map, unpacking, generating a priority queue for adjacent pixel points in descending order of reliability, and then selecting the point with the highest reliability in the priority queue for unpacking each time, and updating the path;

[0078] S305, for each point on the path, correct the jump according to the unwrapped phase of the adjacent pixel points;

[0079] S306, extracting global phase features from the coarse-resolution layer of the unwrapped phase image; refining path selection in the medium-resolution layer, and accurately adjusting local details in the high-resolution layer;

[0080] S307 , smoothing the unwrapped phase image to obtain an unwrapped phase image.

[0081] Specifically, phase unwrapping is performed in the form of multi-resolution reliability sorting of non-continuous dynamic paths, including:

[0082] 1. Input the corrected phase component to obtain the phase diagram;

[0083] 2. Check whether there is severe noise in the input phase image, and pre-smooth the wrapped phase image according to application requirements (optional, such as Gaussian filtering) to reduce the impact of high-frequency noise;

[0084] 3. Calculate reliability index:

[0085] A reliability index is assigned to each pixel to guide subsequent dynamic path planning and unpacking.

[0086] First, calculate the gradient:

[0087] G x (i,j)=φ wrapped (i+1,j)-φ wrapped (i,j) Formula (1)

[0088] G y (i,j)=φ wrapped (i,j+1)-φ wrapped (i,j) Formula (2)

[0089] Among them, G x (i,j) is the horizontal gradient, G y (i,j) is the vertical gradient, φ wrapped is the phase wrapped in the A line sampling point.

[0090] Next, calculate the second-order derivative:

[0091] H x (i,j)=G x (i+1,j)-G x (i,j) Formula (3)

[0092] H y (i,j)=G y (i,h+1)-G y (i,j) Formula (4)

[0093] Among them, H x (i, j) is the second-order derivative in the horizontal direction, H y (i,j) is the second-order derivative in the vertical direction.

[0094] Finally, define the reliability index R(i,j):

[0095]

[0096] Here, ∈ is a positive number less than 10, which is used to avoid the denominator being zero.

[0097] 4. Initialize multi-resolution layering:

[0098] The wrapped phase image is gradually downsampled to generate phase images of different resolutions, for example, by reducing the resolution by a factor of two. The global phase features are then extracted in the coarse resolution layer and the local details are refined in the high resolution layer.

[0099] 5. Dynamic path planning:

[0100] Start unpacking from the point with the highest reliability (such as the center or a specific seed point). Model the unpacking process as a path search problem. Use the A* algorithm to select the neighboring pixel with the highest reliability to expand the path at each step, while avoiding the noise area. Construct a priority queue with reliability R(i,j) as the priority, and select the point with the highest reliability in the queue for unpacking each time. Dynamically update the path to allow jumping between low reliability areas.

[0101] 6. Unpack point by point:

[0102] For each point on the path, the jump is corrected according to the unwrapped phase of the adjacent points:

[0103] This step is divided into two steps: horizontal phase unwrapping and vertical phase unwrapping.

[0104] First, calculate the horizontal phase difference Δφ x (i,j):

[0105] Δφ x (i,j)=φ wrapped (i,j)-φ wrapped (i-1,j) Formula (6)

[0106] Secondly, determine the horizontal phase difference:

[0107] If the horizontal phase difference >π

[0108] φ unwrapped1 (i,j)=φ wrapped (i-1,j)+Δφ x (i,j)-2π Formula (7)

[0109] If the horizontal phase difference <-π

[0110] φ unwrapped1 (i,j)=φ wrapped (i-1,j)+Δφ x (i,j)+2π Formula (8)

[0111] otherwise:

[0112] φ unwrapped1 (i,j)=φ wrapped (i-1,j)+Δφ x (i,j) Formula (9)

[0113] Next, based on the horizontal unwrapped φ unwrapped1 (i,j) performs vertical phase unwrapping.

[0114] Among them, the vertical phase difference Δφ is calculated y (i,j):

[0115] Δφ y (i,j)=φ unwrapped1 (i,j)-φ unwrapped1 (i,j-1) Formula (10)

[0116] If the vertical phase difference >π

[0117] φ unwrapped (i,j)=φ unwrapped1 (i,j-1)+Δφ y (i,j)-2π Formula (11)

[0118] If the vertical phase difference <-π

[0119] φ unwrapped (i,j)=φ unwrapped1 (i,j-1)+Δφ y (i,j)+2π Formula (12)

[0120] otherwise:

[0121] φ unwrapped (i,j)=φ unwrapped1 (i,j-1)+Δφ y (i,j) Formula (13)

[0122] 7. Multi-resolution guided layer-by-layer unwrapping:

[0123] First, the global phase features are extracted from the coarse-resolution layer, then the path selection is refined in the medium-resolution layer and gradually unfolded, and finally the local details are precisely adjusted in the high-resolution layer to smooth the noisy areas.

[0124] 8. Denoising and error control:

[0125] First, the reliability index is used to identify the noisy areas. Then, local filtering or interpolation is applied to the noisy areas to avoid error accumulation. Finally, the unwrapped phase distribution is smoothed in the high-resolution layer.

[0126] 9. Output unwrapping phase

[0127] The unwrapped phase distribution is obtained, which is continuous in the entire image range and reduces the noise effect as much as possible.

[0128] The unwrapping algorithm in the present invention is suitable for phase images with complex structures and high noise, has wider applicability, and can efficiently process large-scale high-resolution phase images and high-noise images. Through the multi-resolution unfolding strategy, the method can also achieve efficient operation in an environment with low computing resources, enhancing the portability and adaptability of the algorithm.

[0129] In terms of performance, this method adopts a two-dimensional phase unwrapping strategy with non-continuous path reliability sorting, and has made multiple optimizations compared to traditional algorithms. Traditional algorithms rely on pixel-by-pixel traversal and global reliability sorting when calculating reliability indicators and non-continuous path searches. This method has extremely large computational and sorting overheads when processing larger-scale phase maps (such as 500x4000 pixels), resulting in slow processing speed. In addition, the path selection of traditional methods is based only on global sorting results, and it is impossible to dynamically avoid noise areas. It is easy to cause error accumulation in high-noise areas, resulting in local errors in the unwrapped phase map.

[0130] To improve efficiency and accuracy, this application introduces sparse matrix and lookup table cache into the algorithm to reduce repeated operations of second-order derivative calculations. At the same time, dynamic path planning is implemented through the A* algorithm and priority queue, replacing the traditional global sorting and significantly shortening the algorithm execution time. In terms of denoising, the dynamic path planning mechanism can more flexibly avoid noisy areas, effectively reduce error accumulation, and improve the reliability of phase unwrapping.

[0131] In terms of accuracy, this method provides global guidance through multi-resolution layered unwrapping, which not only improves the unwrapping accuracy, but also avoids local errors caused by high noise. This multi-resolution strategy processes phase information layer by layer at different resolutions, obtains global phase features at coarse resolution, and refines local details at high resolution, thereby ensuring unwrapping accuracy while controlling time overhead. In addition, the layer-by-layer decomposition characteristics of this strategy help to smooth phase jump areas in high-noise images, thereby improving the denoising effect, allowing the algorithm to still perform well in high-noise environments.

[0132] In summary, the improved algorithm not only has significantly improved the execution speed and processing accuracy, but also has excellent denoising capabilities, and is suitable for efficient unwrapping of large-scale complex phase images.

[0133] In Doppler imaging, phase information is used to determine the direction and speed of blood flow. When the OCT beam encounters a moving scatterer (such as red blood cells in the blood flow), the reflected light changes in frequency, a phenomenon known as the Doppler effect. By analyzing this frequency change, the speed of the scatterer can be calculated.

[0134] Phase is a key parameter in the Doppler signal, which changes as the scatterer moves relative to the probe. Ideally, if the scatterer moves at a constant speed, its phase will change linearly. However, since the phase is periodic, its range is limited to the interval [-π,π] or [0,2π]. This means that when we continuously track the phase change, once the phase exceeds this range, it will "wrap" back into the interval. This phenomenon is called "Phase Wrapping".

[0135] For example, if the phase changes by more than 2π in a sampling period, the observed phase will not increase linearly, but will suddenly jump back to the starting point of the [0,2π] interval, just like the hour hand of a clock suddenly jumps back to 1 o'clock when it exceeds 12 o'clock.

[0136] The phase wrapping phenomenon makes it impossible to obtain continuous velocity information directly from the observed phase data. In order to solve this problem, phase unwrapping is required. The purpose of phase unwrapping is to restore the original linear relationship of phase change, so as to accurately calculate the velocity and displacement of the scatterer.

[0137] Intravascular information is the interference spectrum signal of intravascular tissue collected by intravascular imaging equipment, which is the information obtained by spiral three-dimensional scanning imaging of tissue samples by intravascular imaging equipment. The depth domain complex signal is extracted using the interference spectrum signal, that is, the interference spectrum is Fourier transformed to obtain the depth domain complex signal (including intensity component and phase component) of each spatial point.

[0138] Among them, the intensity component of the complex signal in the depth domain is used to display the structural information of the intravascular tissue to analyze the parameters such as the thickness of the inner, middle and outer membranes of the vascular tissue. The phase component of the complex signal in the depth domain is used to extract the blood flow velocity distribution information and the shear force information of the vascular wall to provide multi-dimensional monitoring information of vascular lesions.

[0139] In the specific implementation, the Doppler velocity of the blood flowing in the blood vessels is extracted by demodulating the phase shift between adjacent A-Lines to account for the rapid movement of blood in the blood vessels. According to the Doppler principle, the frequency shift of the backscattered light of moving particles can be generated, and the OCT complex signal reconstructed by fast Fourier transform can be processed to calculate the phase difference φ between two adjacent A-lines in the B-scan. unwrapped (i, j) to obtain a phase image. This phase difference is caused by the movement of blood in the blood vessels. The flow velocity can be derived from the phase change, as shown in formula (14):

[0140]

[0141] Where Δf(i,j) is the flow velocity, λ is the central wavelength of the light source, n is the refractive index of blood in the tissue (~1.35), τ is the time interval between adjacent A-lines, which is determined by the sweep rate of the light source, and θ is the Doppler angle when measuring the flow velocity, which is determined by the angle between the direction of the catheter focusing beam and the direction of blood flow. In the embodiment, a fixed value of θ=70° is taken. Among them, x and y are the horizontal and vertical coordinates in the polar coordinate image, respectively. Combined with the structure and velocity information obtained by tissue OCT, the high-speed blood flow velocity distribution in the blood vessel is drawn.

[0142] The WSS at the vessel wall and catheter sheath can be obtained by calculating the gradient field of OCT Doppler blood flow distribution. As shown in Formula 12, in a fluid, the shear stress is proportional to the flow velocity gradient, which is given by Newton's viscosity law:

[0143]

[0144] where WSS is the shear stress tensor, μ is the dynamic viscosity of the fluid, and is the flow velocity gradient.

[0145] The velocity gradient It is calculated from the Doppler blood flow distribution image in polar coordinates:

[0146]

[0147] Where du / dx and du / dy are the derivatives of the blood flow velocity components in the x and y directions in polar coordinates.

[0148] It should be noted that in order to facilitate understanding of the embodiments of the present invention, each operation is described as multiple discrete operations. However, the order of description does not represent the order of implementing the operations. In the actual process, the order of the operation steps can be changed as needed. The order of steps in this application is not unique.

[0149] like Figure 3As shown, the present invention also provides an imaging device for analyzing vascular wall shear stress based on intravascular imaging, which is used to implement the steps of the above-mentioned vascular wall shear stress analysis method based on intravascular imaging, including: a swept light source, a reference arm, an imaging device and an image processing module; a part of the swept light emitted by the swept light source enters the reference arm to form a standard signal, and the other part of the light enters the imaging device and irradiates the sample to form a sample signal; the sample signal and the standard signal are interfered to form an IVOCT spectral image, and the spectral image includes a first spectral signal collected under the condition of using a contrast agent and a second spectral signal collected without using a contrast agent; a three-dimensional vascular morphological structure is established with the collected IVOCT first spectral signal by using the image processing module; the vascular wall and artifacts in the second spectral signal are eliminated by using the established three-dimensional vascular morphological structure to obtain a complex signal in the depth domain of the cavity; the complex signal in the depth domain of the cavity is phase unwrapped to obtain an intravascular flow velocity distribution image; the vascular wall shear stress is calculated by using the intravascular flow velocity distribution image to obtain a vascular wall shear stress WSS image.

[0150] Sweep frequency light source 201, 1×2 fiber coupler 202, first fiber circulator 203, reference arm fiber collimator 204, objective lens 205, plane high reflector 206, polarization controller 207, 2×2 fiber coupler 208, photoelectric balance detector 209, signal processing unit 210, second fiber circulator 211, fiber rotation and retraction unit 212, first torque spring 213, first focusing element 214, first reflection element 215, retraction unit 217, second torque spring 218, second focusing element 219, second reflection element 220, distal micro motor 221. Preferably, the center wavelength of the swept frequency light source 201 is 1310nm, the bandwidth is 100nm, and the sweep frequency rate is 100kHz. The first torque spring 213 and the second torque spring 218 are double-layer torque springs.

[0151] The light emitted by the swept frequency light source 201 used in the device of the present invention passes through the 1×2 fiber coupler 202, and then is divided into two sub-beams that enter the first fiber circulator 203 and the second fiber circulator 211 respectively. The light beam that enters the first fiber circulator 203 is transmitted to the reference arm fiber collimator 204 through the polarization controller 207, and then collimated into a parallel light beam by the collimator and irradiated to the plane high reflector 206; the other light beam enters the imaging catheter device and irradiates the sample. In the imaging catheter device, as Figure 2(A) and (B) are two embodiments, which are respectively a proximal drive catheter device and a distal drive catheter device. Among them, the proximal drive catheter device includes: an optical fiber rotation and withdrawal unit 212, which is composed of an optical fiber rotation connector, a rotation motor and a withdrawal linear motor, and the torque and tension output by the motor are transmitted to the distal end of the catheter through the first torque spring 213 to perform spiral scanning, and the first focusing element 214 in the distal end of the catheter converges the light beam through the first reflective element 215 to turn the focused light to the first sample 216 on the side of the catheter; the distal drive catheter device includes: a withdrawal unit 217 composed of a withdrawal linear motor, a second focusing element 219 in the distal end of the catheter converges the light beam through the second reflective element 220 to turn the focused light to the second sample 222 on the side of the catheter, and the tension output by the motor is transmitted to the distal end of the catheter through the second torque spring 218, and the distal end of the catheter The micro motor 221 rotates the second reflective element 220 to perform spiral scanning.

[0152] The vessel wall is spirally scanned, and 3D reconstruction of the vessel is achieved by simultaneously rotating the drive shaft and axially translating the catheter probe.

[0153] Figure 5 The figure shows the OCT structure diagram and blood flow velocity distribution diagram obtained by testing in a phantom according to the method disclosed in the present invention. Figure 3 The sweep frequency OCT imaging device and imaging catheter based on the proximal drive catheter are shown to image a vascular phantom (inner diameter 2.5mm, outer diameter 4mm). During the imaging process, a milk solution is used to simulate blood. The injection pump is turned on and a 10mL syringe is used to pump the solution at a preset injection speed to simulate the blood flow in the cardiovascular cavity. A single in-situ scan is performed on the same cross-sectional position of the vascular phantom, that is, the imaging scanning probe only performs a rotational scanning motion at a predetermined position and does not retract, so as to obtain the corresponding OCT signal, and the blood flow velocity distribution algorithm shown in the present invention is used to obtain the blood flow velocity distribution image in the blood vessel. Figure 5 (a) is a structural image of a certain cross section of the vascular phantom. The vascular phantom, milk and various reflective surfaces can be clearly distinguished by the signal intensity. Figure 5 (b), (c) and (d) are the velocity distribution images of the blood vessel cavity in this section before phase correction, after phase correction and after phase unwrapping, respectively. The velocity distribution has been effectively extracted and corrected.

[0154] Figure 6 The figure shows the local average flow velocity curve obtained by using the present invention. The experimental setting is divided into 5 groups, and the flow velocities in the phantom are converted to 500mm / s, 400mm / s, 300mm / s, 200mm / s and 100mm / s respectively. The local average flow velocity curve shows that different Doppler flow velocity distributions can be measured in the phantoms with different flow velocities. The flow velocities in the experimental setting have a linear relationship, and the measured Doppler flow velocity distributions also have a corresponding linear relationship.

[0155] Obviously, the above embodiments are only examples for clear explanation, and are not intended to limit the implementation methods. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from them are still within the protection scope of the invention.

Claims

1. A method for analyzing vascular wall shear stress based on intravascular imaging, characterized in that: The following steps are involved: S1, collecting IVOCT spectral signals in two groups of target blood vessels; S2, using the three-dimensional vascular morphology structure established by the first group of spectral signals, removing the vascular wall and artifacts in the second group of spectral signals, and obtaining a complex signal in the cavity depth domain; S3, performing Doppler frequency shift calculation on the intracavitary depth domain complex signal and performing phase unwrapping to obtain an intravascular flow velocity distribution image; S4. Calculate the vascular wall shear stress using the intravascular flow velocity distribution image to obtain the vascular wall shear stress WSS image.

2. The method for analyzing vascular wall shear stress based on intravascular imaging according to claim 1, characterized in that: In S1, two sets of IVOCT spectral signals in target blood vessels are collected, including: Process 1: Place the imaging catheter at a predetermined position in the target blood vessel, flush an appropriate amount of saline or contrast agent into the blood vessel, and the catheter starts to rotate. When there is no blood obstruction in real-time imaging, the catheter is withdrawn to obtain the first spectral signal during the withdrawal process; Process 2: The imaging catheter is placed again at a predetermined position in the target blood vessel, the catheter starts to rotate, an IVOCT spectral image is acquired, an electrocardiogram signal is acquired synchronously, and all the acquired data within one heartbeat cycle is acquired as a second spectral signal.

3. The method for analyzing vascular wall shear stress based on intravascular imaging according to claim 2, characterized in that: In S2, the three-dimensional vascular morphology structure established by the first set of spectral signals is used to remove the vascular wall and artifacts in the second set of spectral signals to obtain a complex signal in the cavity depth domain, including the following steps: S201, performing Fourier transform on the second spectral image to obtain a depth domain complex signal of IVOCT; S202, extracting intensity components from the depth domain complex signal based on IVOCT; S203, using intensity components to divide the structural information of the vascular tissue to obtain a structural image, aligning the structural image with the three-dimensional vascular morphological structure, and removing the vascular wall and artifacts in the structural image using the three-dimensional vascular morphological structure to obtain an intracavitary structural image and an intracavitary depth domain complex signal.

4. The method for analyzing vascular wall shear stress based on intravascular imaging according to claim 3, characterized in that: After S203, the method further includes subtracting the phase information of the inner and outer walls of the sheath or the self-coherent signal inside the system from the phase component extracted from the intracavity depth domain complex signal to obtain a corrected phase component.

5. The method for analyzing vascular wall shear stress based on intravascular imaging according to any one of claim 4, characterized in that: In S3, the Doppler frequency shift calculation is performed on the intracavity depth domain complex signal, and phase unwrapping is performed, including: S301, using the phase component of the complex signal in the cavity depth domain after correction, calculate the Doppler frequency shift to obtain a wrapped phase map; S302, calculating a reliability index according to the second-order derivative of the wrapped phase of each pixel in the wrapped phase image; S303, gradually downsampling the wrapped phase image to generate phase images with different resolutions; S304, starting from the point with the highest reliability in the phase map, unpacking, generating a priority queue for adjacent pixel points in descending order of reliability, and then selecting the point with the highest reliability in the priority queue for unpacking each time, and updating the path; S305, for each point on the path, correct the jump according to the unwrapped phase of the adjacent pixel points; S306, extracting global phase features from the coarse-resolution layer of the unwrapped phase image; refining path selection in the medium-resolution layer, and accurately adjusting local details in the high-resolution layer; S307 , smoothing the unwrapped phase image to obtain an unwrapped phase image.

6. The method for analyzing vascular wall shear stress based on intravascular imaging according to claim 5, characterized in that: In S302, the reliability index is calculated according to the following formula: Among them, H x (i, j) is the horizontal second-order derivative of pixel (i, j), H y (i, j) is the second-order derivative of pixel (i, j) in the vertical direction, R(i, j) is the reliability index, and ∈ is a positive number less than 10.

7. The method for analyzing vascular wall shear stress based on intravascular imaging according to claim 5, characterized in that: In S3, the phase difference between any two adjacent A-lines is calculated from the unwrapped phase signal. A phase image is obtained, and the intravascular flow velocity is calculated according to the phase image according to the following formula: Where Δf (i，j) is the flow velocity, λ is the central wavelength of the light source, n is the refractive index of blood in the tissue, τ is the time interval between adjacent A-lines, θ is the Doppler angle when measuring the flow velocity, which is determined by the angle between the focusing beam direction of the catheter and the blood flow direction, i and j are the horizontal and vertical coordinates in the polar coordinate image respectively; φ unwrapped (i, j) is the phase difference between two adjacent A-lines.

8. The method for analyzing vascular wall shear stress based on intravascular imaging according to claim 1, characterized in that: In S4, the blood vessel wall shear stress is calculated using the blood vessel flow velocity distribution image, including: according to the fact that the shear stress is proportional to the flow velocity gradient in the fluid, the shear stress is calculated according to the following formula: Where du / dx and du / dy are the derivatives of the blood flow velocity components in the x and y directions in polar coordinates.

9. An imaging device for analyzing vascular wall shear stress based on intravascular imaging, used to implement the steps of the method for analyzing vascular wall shear stress based on intravascular imaging as described in any one of claims 1 to 8, comprising: A swept frequency light source, a reference arm, an imaging device and an image processing module; a part of the swept frequency light emitted by the swept frequency light source enters the reference arm to form a standard signal, and another part of the light enters the imaging device and irradiates the sample to form a sample signal; the sample signal and the standard signal are interfered to form an IVOCT spectral image; the spectral image includes a first spectral signal collected under the condition of using a contrast agent and a second spectral signal collected without using a contrast agent; the image processing module uses the three-dimensional vascular morphological structure established by the collected first spectral signal to eliminate the vascular wall and artifacts in the second spectral signal to obtain an intracavitary depth domain complex signal; the phase component of the intracavitary depth domain complex signal is phase unwrapped to obtain an intravascular flow velocity distribution image; the intravascular flow velocity distribution image is used to calculate the vascular wall shear stress to obtain a vascular wall shear stress WSS image.

10. The imaging device for analyzing vascular wall shear stress based on intravascular imaging according to claim 8, characterized in that: The imaging catheter device includes a proximal drive catheter device or a distal drive catheter device; The proximal drive catheter device includes: an optical fiber rotation and retraction unit, which is composed of an optical fiber rotation connector, a rotation motor and a retraction linear motor. The torque and tension output by the motor are transmitted to the distal end of the catheter through a first torque spring to perform spiral scanning. The first focusing element in the distal end of the catheter converges the light beam, and the focused light is turned to the first sample on the side of the catheter through a first reflection element. The distal drive catheter device includes: a retraction unit; the retraction unit is composed of a retraction linear motor, a second focusing element in the distal end of the catheter converges the light beam through the second reflecting element to turn the focused light to the second sample on the side of the catheter, and the tension output by the linear motor is transmitted to the distal end of the catheter through the second torque spring, and the second reflecting element is rotated to perform spiral scanning in cooperation with the micromotor at the distal end of the catheter.

Images