Blood vessel wall thickness measuring method based on coronary artery reconstruction

Abstract

The invention belongs to the technical field of medical image processing, and particularly relates to a blood vessel wall thickness measuring method based on coronary artery reconstruction. According to the technical scheme, based on multi-row single-heartbeat CT imaging, the optimal cardiac cycle is selected through Laplacian smooth mapping, image artifacts are eliminated through a U-net architecture with expansion convolution, the coronary artery is reconstructed, segmentation of the blood vessel wall and the lumen is conducted on the reconstructed coronary artery, and the thickness of the blood vessel wall is measured through a maximum internally tangent ball method. The problems that coronary artery imaging has artifacts caused by heart movement, adjacent imaging is discontinuous and accurate reconstruction cannot be achieved due to vasoconstriction are solved, and the problem of selecting the optimal period of coronary vessel wall thickness measurement is solved.

Description

Technical Field

[0001] The invention belongs to the technical field of medical image processing, and in particular is a method for measuring the thickness of a blood vessel wall based on coronary artery reconstruction. Background Art

[0002] The thickness of the coronary artery wall refers to the thickness of the intima, media and adventitia of the coronary artery, and is an important indicator for assessing the risk of cardiovascular disease. Typical cardiovascular diseases include atherosclerosis and hypertension. Atherosclerosis is a chronic inflammatory disease characterized by the accumulation of lipid plaques in the arterial wall. These plaques may thicken and calcify over time, causing the blood vessel wall to thicken and eventually lead to vascular stenosis; hypertension is a phenomenon in which the pressure on the blood vessel wall caused by blood flowing is continuously higher than normal, and the blood vessel wall is in a state of high pressure for a long time, and the wall is stretched and thinned. These diseases are characterized by being difficult to detect before onset and difficult to screen early, but they will cause micro-deformations in the thickness of the blood vessel wall. Therefore, it is necessary to measure the thickness of the coronary artery wall.

[0003] The prerequisite for measuring the thickness of the coronary artery wall is to obtain coronary artery imaging and then segment the captured image. The existing coronary artery imaging methods include coronary CT, coronary angiography, vascular ultrasound, etc. At present, since the beating of the heart will have a significant impact on coronary artery imaging, resulting in image artifacts, no matter what imaging method is used, in order to obtain complete, artifact-free coronary artery imaging, the subsequent processing ideas of the image are generally the same, which are divided into two steps: artifact elimination and image reconstruction.

[0004] The patent with the publication number of "CN117064446A Dynamic Three-Dimensional Reconstruction System of Vascular Vessels Based on Intravascular Ultrasound" provides a method for dynamic reconstruction of cardiac vessels based on intravascular ultrasound imaging. This method performs three-dimensional reconstruction on two-dimensional images obtained by intravascular ultrasound imaging, establishes an artifact elimination network to extract accurate vascular walls, and uses a generative adversarial network to compensate for frame loss in the intravascular ultrasound sequence after extracting key frames, and finally realizes dynamic three-dimensional reconstruction of cardiac vessels based on the cardiac motion cycle. However, the changes in vascular diameter and vascular wall thickness caused by vascular contraction under cardiac motion lead to different coronary vascular wall thickness at the same position captured at different times in imaging, which brings great difficulties to the selection of vascular wall thickness measurement time. The selection of the optimal cardiac cycle and the optimal phase is of great significance in the measurement of vascular wall thickness. This method includes multiple cardiac cycles in one intravascular ultrasound imaging, and aligns the lumen area of ​​the vascular contraction period and the expansion period under different cardiac cycles based on the gated frame, but ignores the consideration of vascular reconstruction under the optimal cardiac cycle. The vascular wall thickness obtained by segmentation after vascular reconstruction is dynamic and cannot accurately reflect the micro-deformation of the blood vessel. Summary of the invention

[0005] In view of the above problems, the present invention proposes a method for measuring the vascular wall thickness based on coronary artery reconstruction, which can eliminate artifacts, solve the influence of lumen area caused by vascular contraction, reconstruct the coronary artery in the optimal cardiac cycle, and measure the coronary artery wall thickness, thereby improving the accuracy of the results.

[0006] The technical solution of the present invention is:

[0007] A method for measuring the thickness of a blood vessel wall based on coronary artery reconstruction comprises the following steps:

[0008] S1. Use multi-row single heartbeat CT imaging to collect images of specified phases in multiple cardiac cycles within a set time. Define a set of images taken by multi-row single heartbeat CT imaging as C, so as to obtain image data sets at different phases at various positions of the coronary artery. Define the image sequence obtained at the coronary artery position c as V c ;

[0009] S2. Select the single heartbeat CT imaging of the optimal adjacent phase according to the collected data set, specifically:

[0010] The image sequence V c Mapping from high-dimensional space to low-dimensional space, and then selecting the cluster center through clustering algorithm, obtain the cluster center distance of CT imaging at the coronary artery position c in adjacent phases, which is defined as Then, according to the number of images taken by a multi-row single heartbeat CT imaging, the average value of the cluster center distance of each position of the coronary artery in adjacent phases is obtained, which is expressed as Finally, the phase that achieves the best imaging effect in all sequences is obtained Expressed as K represents the number of cardiac cycles;

[0011] S3. Construct an image artifact removal network to reconstruct the coronary arteries, specifically:

[0012] The image artifact removal network includes a deformable U-net, a scaling layer, and a spatial transformation network; the input of the deformable U-net is the forward sequence and reverse sequence of the image, based on the obtained phase In chronological order from each V c Extraction in phase The images captured at the same time are obtained as continuous frames. So as to obtain the forward sequence And record it as continuous frame F, and get the reverse sequence And recorded as continuous frame R; F and R are extracted by deformable U-net, and then the segmented image is obtained by 1x1x1 convolution and activation with a stride of 1. and The velocity field is obtained by a 5x5x5 convolution with a stride of 1. and Velocity Field and After scaling the layer, we get the deformation field and The deformation field and the segmentation results are input into the spatial transformation network to obtain the image sequence after the artifacts are removed. Finally, the image sequence after the artifacts are removed is reconstructed in three dimensions through the reconstruction network to obtain the three-dimensional image of the coronary artery;

[0013] S4. Using the obtained three-dimensional image of the coronary arteries, the maximum endosphere method is used to measure the thickness of the vascular wall.

[0014] Furthermore, 128-row single heartbeat CT imaging was used, the time was set to 12-15s, 20 cardiac cycles were collected, and the preset phase interval between each two single heartbeat CT imaging was 3%, and images within the phase range of 30%-60% of the systolic period and 60%-90% of the diastolic period were collected.

[0015] Furthermore, the image sequence V is transformed into c Mapping from high-dimensional space to low-dimensional space, the objective function of the Laplace eigenmap optimization is:

[0016]

[0017] Among them, V c As a set of data instances, y i ,y j is the vector representation of i and j in the data instance in the m-dimensional subspace, W ij The adjacency matrix constructed

[0018]

[0019] Here, σ is a specified constant.

[0020] Furthermore, the method for obtaining the cluster center distance is:

[0021] For any phase n, the sum of its distances to the cluster centers of the adjacent phases n-1 and n+1 is defined as

[0022]

[0023] in It represents the distance between the cluster centers of the images at the same position c of the coronary artery captured at the phases of 30%+(n-1)×3% and 30%+n×3%.

[0024] Furthermore, the deformable U-net network includes a five-layer deformable U-net, whose encoding stage is a contraction path and the decoding stage is an expansion path; there is a concatenation in the contraction path, merging the features extracted from the normal convolution and the dilated convolution, and in the expansion path the feature map is upsampled to the original resolution of the input image for inter-frame registration and segmentation; a concatenation is used between the contraction path and the expansion path to feed the high-resolution features into the upsampled output.

[0025] The two sequence images are segmented in sequence through a 1x1x1 convolution with a stride of 1 and an activation function to obtain the segmented image and In order to preserve both global shape similarity and boundary details, the segmentation loss adopts a hybrid segmentation loss combined with class-weighted Tversky loss. and class-weighted cross entropy loss

[0026]

[0027] in and is the true value segmentation containing some frame annotations, gt represents the ground truth, Calculated on the labeled voxels;

[0028] The pixel velocity field is obtained by a 5x5x5 convolution with a stride of 1. and That is, the displacement direction and movement speed of each pixel are obtained. and In the definition and are the velocity fields extracted from the image frames obtained at the same position c when they are input in positive order and reverse order respectively;

[0029] Velocity Field and After scaling the layer, we get the deformation field and definition is the image frame obtained at position c in the continuous frames F and R, and and Respectively represent the artifact removal arrive The deformation field and arrive The deformation field of ; the pixel smoothing loss is defined as:

[0030]

[0031] Where N is the number of pixels, i represents the pixel, Deformation Field The deformation field at pixel i in , where x and y represent the position of pixel i;

[0032] Furthermore, the deformation field and the segmented image are processed through a spatial transformation network to obtain a de-artifacted image sequence, including a forward sequence and reverse sequence This image series has been segmented into membrane boundaries, including the inner and outer membranes;

[0033] In the spatial transformer network, the deformation field is used to segment the image and Perform spatial transformation:

[0034]

[0035] Where ◎ represents pixel multiplication, and Respectively and In the segmented image at position c, and Represents consecutive frames and The image frame obtained at position c in .

[0036] Define the loss function for reconstructing artifact-free image sequences

[0037]

[0038] Among them, u∈{0,1,2} represents three regions, namely, the region outside the vessel wall, the region between the vessel wall and the intima boundary, and the region inside the intima lumen. Ω(u≠0) represents all regions except the vessel wall, p represents the voxel belonging to the Ω(u≠0) region in the image frame, and R i (p) represents the image intensity at voxel p in the i-th frame in the continuous frame R, F i (p) represents the image intensity at voxel p in the i-th frame in the continuous frame F, Reconstruction of continuous frames The image intensity at voxel p in the i-th frame is, Reconstruction of continuous frames The image intensity at voxel p in the i-th frame is, and They represent the average image intensity of the i-th frame in the corresponding continuous frames;

[0039] Furthermore, the method for reconstructing 3D coronary arteries is as follows: the artifact-free forward sequence is passed through the reconstruction network, stacked along the membrane boundary along the coronary CT acquisition direction, linear interpolation is performed on the continuous 2D contours, and then the 3D surface is visualized by the marching cubes algorithm;

[0040] The total loss function of the network is:

[0041]

[0042] Among them, λ1, λ2 and λ3 are the parameters of various types of losses respectively. The network parameters are obtained by pre-training the network through the collected data based on the loss function.

[0043] Further, the specific method of measuring the thickness of the blood vessel wall using the maximum inscribed sphere method is:

[0044] The maximum endosphere method was used to make the centerline of the coronary artery and lumen. tol and l q , calculate the thickness of the blood vessel wall at each position, define the same plane perpendicular to the entrance of this section of blood vessels and the two midlines l tol and l q The intersection points are o tol and q The radii of the inscribed planes are r tol and r q , with o tol and q The direction of the line is the x-axis, o q Take O as the origin and make a plane coordinate system. The thickness of the blood vessel wall at any point on the plane is m i for:

[0045]

[0046] where θ i is the angle between the line from the inscribed plane i to the origin and the x-axis, and r is the midline l of the two planes tol and l q Intersection o tol and q The straight-line distance.

[0047] The beneficial effects of the present invention are that, compared with the prior art, the technical solution proposed by the present invention is based on multi-row single heartbeat CT imaging, uses Laplace smoothing mapping to select the best cardiac cycle, uses a U-net architecture with dilated convolution to eliminate image artifacts, and performs segmentation of the vascular wall and lumen, reconstructs the coronary artery through the reconstruction network, and uses the maximum inscribed sphere method to measure the vascular wall thickness. The problem of artifacts in coronary artery imaging caused by cardiac motion, discontinuity of adjacent imaging caused by vascular contraction, and inability to accurately reconstruct is solved, as well as the problem of selecting the best period for measuring the coronary vessel wall thickness. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is the overall flow chart of the present invention.

[0049] Figure 2Schematic diagram of the structure of the coronary artery reconstruction model of the present invention.

[0050] Figure 3 Schematic diagram of the deformable U-net structure.

[0051] Figure 4 Schematic diagram of calculating blood vessel wall thickness using the maximum inscribed sphere method. DETAILED DESCRIPTION

[0052] The technical principle and scheme of the present invention are described in detail below with reference to the accompanying drawings:

[0053] like Figure 1 As shown, the present invention mainly comprises the following steps:

[0054] 1) Original data set collection, specifically:

[0055] Use 128-row single-heartbeat CT imaging to collect 20 cardiac cycles, lasting about 12-15 seconds. The preset phase interval between each two single-heartbeat CT imaging is 3%, and images are collected within the phase range of 30%-60% of the systolic phase and 60%-90% of the diastolic phase.

[0056] The number of images captured by a multi-slice single heartbeat CT imaging is denoted as C. The images captured at different phases at different locations of the coronary artery are represented as Where V c is a sequence of images taken at the coronary artery position c in time sequence (i.e., at 20 different phases),

[0057] 2) According to the collected data set, the single heartbeat CT imaging of the optimal adjacent phase is selected, specifically:

[0058] Input image sequence V c , Laplace smoothing mapping is used to map the real image from high-dimensional space to low-dimensional space.

[0059] The Laplacian eigenmap reconstructs the local structural features of the data manifold by constructing a graph with an adjacency matrix W. The main idea is that if two data instances i and j are very similar, then i and j should be as close as possible in the target subspace after dimensionality reduction. c As a set of data instances. Assume that the number of data instances is n, and the dimension of the target subspace, i.e. the final dimension reduction target, is m. Define a matrix Y of size n×m, where each row vector y i is the vector representation of data instance i in the target m-dimensional subspace (i.e., data instance i after dimensionality reduction). Our goal is to make similar data instances i and j as close as possible in the target subspace after dimensionality reduction, so the objective function of Laplace eigenmap optimization is as follows:

[0060] min∑ i ||y i -y j || 2 W ij (1)

[0061] Among them yi, y j is the vector representation of data instances i and j in the m-dimensional subspace, Wi j The adjacency matrix constructed is expressed as:

[0062]

[0063] Here, σ is a specified constant.

[0064] After obtaining the low-dimensional feature vector, the AP clustering algorithm (Affinity propagation) is used to select the final cluster center.

[0065] To compare the input image sequence V c The positions of the cluster centers of adjacent images in the input image sequence V c Medium Image The two adjacent phase images and The sum of the distances to the cluster centers is where c∈[1,C], n∈[1,20].

[0066] For any phase n, take the adjacent phases n-1, n and n+1 (except when n=1 or n=20) before and after it, and record the cluster center distance sequence of the CT imaging corresponding to each position of the coronary artery in this group of adjacent phases as

[0067]

[0068] in It represents the distance between the cluster centers of the images at the same position c of the coronary artery captured at the phases of 30%+(n-1)×3% and 30%+n×3%.

[0069] Calculate the average distance between the cluster centers of each coronary artery position in each adjacent phase

[0070]

[0071] Where C is the number of images captured in a multi-row single heartbeat CT imaging. is the cluster center distance of CT imaging at the coronary artery position c in adjacent phases.

[0072] Comprehensively select the phase that achieves the best imaging effect (i.e., the least artifact effect) among all sequences Right now

[0073] 3) Build Figure 2 The image artifact removal network shown is used to reconstruct the coronary arteries, specifically:

[0074] (1) From each V c Extraction in phase The images captured at the same time are obtained as continuous frames. The forward sequence Recorded as continuous frame F (Forward Image), the reverse sequence Recorded as continuous frame R (Reverse Image).

[0075] In the forward sequence In the c-th frame, the phase The image captured at position c is recorded as In reverse sequence In the cth frame, the phase The image taken at position (C-c+1) is recorded as However, whether in F or R, The meanings are the same, both indicating the phase Image taken at position c at .

[0076] (2) The above continuous frames are respectively input into a five-layer deformable U-net network, that is, dilated convolution is introduced into the U-net to obtain a larger receptive field, and the velocity field and segmentation image are finally obtained through inter-frame registration and segmentation.

[0077] like Figure 3 As shown in Figure 1, the deformable U-net architecture has two main paths: a contraction path in the encoding stage and an expansion path in the decoding stage. In the contraction path, there is a concatenation that merges the features extracted from the normal convolution and the dilated convolution to extract features of different scales. In the expansion path, the feature map is upsampled to the original resolution of the input image for inter-frame registration and segmentation. In addition, a concatenation is used between the contraction path and the expansion path to feed the high-resolution features into the upsampled output, thereby improving accuracy.

[0078] At the end of the deformable U-net architecture decoder, a 5x5x5 convolution with a stride of 1 is performed to obtain the velocity field of the pixel. and That is, the displacement direction and movement speed of each pixel are obtained. and middle, and They are the velocity fields extracted from the image frames obtained at the same position c when they are input in positive and reverse order.

[0079] At the same time, at the end of the deformable U-net architecture decoder, the two sequence images are segmented in turn through a 1x1x1 convolution with a stride of 1 to obtain the segmented image and In order to preserve both global shape similarity and boundary details, a hybrid segmentation loss combining class-weighted Tversky loss is adopted. and class-weighted cross entropy loss

[0080]

[0081] in and is the true segmentation containing partial frame annotations, and gt represents the ground truth. Computed over labeled voxels.

[0082] (3) Velocity field and After scaling the layer, we get the deformation field and definition is the image frame obtained at position c in the continuous frames F and R, and and Respectively represent the artifact removal arrive The deformation field and artifact removal arrive The deformation field of ; the pixel smoothing loss is defined as:

[0083]

[0084] Where N is the number of pixels, i represents the pixel, Deformation Field The deformation field at pixel i in , where x and y represent the position of pixel i.

[0085] (4) The deformation field and segmented image are used to obtain the de-artifacted image sequence through a spatial transformation network.

[0086] The deformation field and the segmented image are input into the spatial transformation network to obtain the image sequence after artifact removal, including the forward sequence and reverse sequence The image series has been segmented with membrane boundaries, including the inner membrane and the outer wall.

[0087] In the spatial transformer network, the deformation field is used to segment the image and Perform spatial transformation:

[0088]

[0089] Where ◎ represents pixel multiplication, and Respectively and In the segmented image at position c, and Represents consecutive frames and The image frame obtained at position c in .

[0090] In this method, the Normalized Cross Correlation (NCC) principle is used to measure the similarity between the images reconstructed from the forward sequence and the reverse sequence.

[0091]

[0093] Among them, u∈{0,1,2} represents three regions, namely, the region outside the vessel wall, the region between the vessel wall and the intima boundary, and the region inside the intima lumen. Ω(u≠0) represents all regions except the vessel wall, p represents the voxel belonging to the Ω(u≠0) region in the image frame, and R i (p) represents the image intensity at voxel p in the i-th frame in the continuous frame R, F i (p) represents the image intensity at voxel p in the i-th frame in the continuous frame F, Reconstruction of continuous frames The image intensity at voxel p in the i-th frame is, Reconstruction of continuous frames The image intensity at voxel p in the i-th frame is, and They respectively represent the average image intensity of the i-th frame in the corresponding continuous frames.

[0094] Get the loss function of the reconstructed artifact-free image sequence

[0095]

[0096] (5) Use the reconstruction network to perform three-dimensional reconstruction of the coronary arteries based on the forward artifact removal sequence.

[0097] The reconstruction network is as follows: the membrane boundaries are stacked along the coronary CT acquisition direction. Linear interpolation is performed on the continuous 2D contours, and then the 3D surface is visualized by the marching cubes algorithm.

[0098] Finally, the total loss function for coronary artery reconstruction is:

[0099]

[0100] Among them, λ1, λ2 and λ3 are the parameters of each type of loss respectively.

[0101] 4) Measure the thickness of the vascular wall using the maximum inscribed sphere method, specifically:

[0102] The maximum endosphere method was used to make the centerline of the coronary artery and lumen. tol and l q Calculate the thickness of the blood vessel wall at each location. The same plane perpendicular to the entrance of this section of the blood vessel and the two midlines l tol and l q The intersection points are o tol and q The radii of the inscribed planes are r tol and r q , with o tol and q The direction of the line is the x-axis, o q Take O as the origin and make a plane coordinate system. The thickness of the blood vessel wall at any point on the plane is m i for:

[0103]

[0104] where θ i is the angle between the line from the inscribed plane i to the origin and the x-axis, and r is the midline l of the two planes tol and l q Intersection o tol and q The straight-line distance.

[0105] In summary, the present invention has the following characteristics and advantages:

[0106] The present invention collects coronary artery images of 20 cardiac cycles through 128-row single-heartbeat CT imaging, selects the optimal cardiac cycle using Laplace smoothing mapping, constructs an image artifact elimination network, and performs segmentation of the vascular wall and the lumen. The coronary artery is reconstructed through the reconstruction network, and the maximum inscribed sphere method is used to measure the vascular wall thickness.

[0107] Due to the contraction of blood vessels caused by cardiac movement, the lumen area and vascular wall thickness at the same position are different at different phases, resulting in the coronary artery images at adjacent positions taken at different phases in a single imaging are not smooth and continuous. Therefore, the 128-row single heartbeat CT imaging technology was selected to solve the problem that different positions in coronary artery imaging cannot be imaged simultaneously, and solve the problem of acquisition difficulty caused by the imaging time.

[0108] Since cardiac motion can significantly affect coronary artery imaging when using CT imaging, resulting in image artifacts, it is necessary to build an image artifact elimination network to select the optimal cardiac cycle for coronary artery reconstruction. The optimal phase is selected using Laplace smoothing mapping, and detailed features are extracted using a U-net architecture with dilated convolution, a deformation field is constructed, and the coronary artery is reconstructed at the optimal phase time.

[0109] For the segmented coronary arteries, the maximum inscribed sphere method was used to obtain the coronary artery as a whole and the lumen midline, and the maximum inscribed sphere plane was used to calculate the vascular wall thickness at any point of the coronary artery.

Claims

1. A method for measuring vascular wall thickness based on coronary artery reconstruction, characterized in that: The following steps are involved: S1. Use multi-row single heartbeat CT imaging to collect images of specified phases in multiple cardiac cycles within a set time. Define a set of images taken by multi-row single heartbeat CT imaging as C, so as to obtain image data sets at different phases at various positions of the coronary artery. Define the image sequence obtained at the coronary artery position c as V c ; S2. Select the single heartbeat CT imaging of the optimal adjacent phase according to the collected data set, specifically: The image sequence V c Mapping from high-dimensional space to low-dimensional space, and then selecting the cluster center through clustering algorithm, obtain the cluster center distance of CT imaging at the coronary artery position c in adjacent phases, which is defined as Then, according to the number of images taken by a multi-row single heartbeat CT imaging, the average value of the cluster center distance of each position of the coronary artery in adjacent phases is obtained, which is expressed as Finally, the phase that achieves the best imaging effect in all sequences is obtained Expressed as K represents the number of cardiac cycles; S3. Construct an image artifact removal network to reconstruct the coronary arteries, specifically: The image artifact removal network includes a deformable U-net, a scaling layer, and a spatial transformation network; the input of the deformable U-net is the forward sequence and reverse sequence of the image, based on the obtained phase In chronological order from each V c Extraction in phase The images captured at the same time are obtained as continuous frames. So as to obtain the forward sequence And record it as continuous frame F, and get the reverse sequence And recorded as continuous frame R; F and R are extracted by deformable U-net, and then the segmented image is obtained by 1x1x1 convolution and activation with a stride of 1. and The velocity field is obtained by a 5x5x5 convolution with a stride of 1. and Velocity Field and After scaling the layer, we get the deformation field and The deformation field and the segmentation results are input into the spatial transformation network to obtain a sequence of images after artifact removal; finally, the sequence of images after artifact removal is passed through the reconstruction network to perform three-dimensional reconstruction of the coronary arteries to obtain a three-dimensional image of the coronary arteries; S4. Using the obtained three-dimensional image of the coronary arteries, the maximum endosphere method is used to measure the thickness of the vascular wall.

2. The method for measuring vascular wall thickness based on coronary artery reconstruction according to claim 1, characterized in that: In S1, 128-row single-heartbeat CT imaging is used, the time is set to 12-15s, 20 cardiac cycles are collected, and the preset phase interval between each two single-heartbeat CT imaging is 3%, and images within the phase range of 30%-60% of the systolic period and 60%-90% of the diastolic period are collected.

3. The method for measuring vascular wall thickness based on coronary artery reconstruction according to claim 2, characterized in that: In S2, the image sequence V is transformed into c Mapping from high-dimensional space to low-dimensional space, specifically the objective function of the Laplace eigenmap optimization is: Among them, V c As a set of data instances, y i ,y j is the vector representation of i and j in the data instance in the m-dimensional subspace, W ij The adjacency matrix constructed Here, σ is a specified constant.

4. The method for measuring vascular wall thickness based on coronary artery reconstruction according to claim 2, characterized in that: In S2, the cluster center distance is obtained as follows: For any phase n, the sum of the distances between it and the cluster centers of the adjacent phases n-1 and n+1 is defined as A c n : in It represents the distance between the cluster centers of the images located at the same position c of the coronary artery captured at the phases of 30%+(n-1)×3% and 30%+n×3%.

5. The method for measuring vascular wall thickness based on coronary artery reconstruction according to claim 1, characterized in that: In S3, the deformable U-net network includes five layers of deformable U-net, whose encoding stage is a contraction path and the decoding stage is an expansion path; there is a series connection in the contraction path, which combines the features extracted from the normal convolution and the dilated convolution, and in the expansion path, the feature map is upsampled to the original resolution of the input image for inter-frame registration and segmentation; a series connection is used between the contraction path and the expansion path to feed the high-resolution features into the upsampled output; The two sequence images are segmented in sequence through a 1x1x1 convolution with a stride of 1 and an activation function to obtain the segmented image and The segmentation loss uses a hybrid segmentation loss combined with class-weighted Tversky loss and class-weighted cross entropy loss in and is the true value segmentation containing some frame annotations, gt represents the ground truth, Calculated on the labeled voxels; The pixel velocity field is obtained by a 5x5x5 convolution with a stride of 1. and That is, obtain the displacement direction and movement speed of each pixel; in the velocity field and In the definition and are the velocity fields extracted from the image frames obtained at the same position c when they are input in positive order and reverse order respectively; Velocity Field and After scaling the layer, we get the deformation field and definition is the image frame obtained at position c in the continuous frames F and R, and and Respectively represent the artifact removal arrive The deformation field and arrive The deformation field of ; the pixel smoothing loss is defined as: Where N is the number of pixels, i represents the pixel, Deformation Field The deformation field at pixel i in , where x and y represent the position of pixel i; In the spatial transformer network, the deformation field is used to segment the image and Perform spatial transformation to obtain the image sequence after artifact removal: Where ◎ represents pixel multiplication, and Respectively and In the segmented image at position c, and Represents consecutive frames and The image frame obtained at position c in ; Define the loss function for reconstructing artifact-free image sequences Among them, u∈{0,1,2} represents three regions, namely, the region outside the vessel wall, the region between the vessel wall and the intima boundary, and the region inside the intima lumen. Ω(u≠0) represents all regions except the vessel wall, p represents the voxel belonging to the Ω(u≠0) region in the image frame, and R i (p) represents the image intensity at voxel p in the i-th frame in the continuous frame R, F i (p) represents the image intensity at voxel p in the i-th frame in the continuous frame F, Reconstruction of continuous frames The image intensity at voxel p in the i-th frame is, Reconstruction of continuous frames The image intensity at voxel p in the i-th frame is, and They represent the average image intensity of the i-th frame in the corresponding continuous frames; The specific method for reconstructing 3D coronary arteries is: the artifact-free forward sequence passes through the reconstruction network, stacks along the membrane boundary along the coronary CT acquisition direction, performs linear interpolation on continuous 2D contours, and then visualizes the 3D surface through the marching cube algorithm; The final network total loss function is: Among them, λ1, λ2 and λ3 are the parameters of various types of losses respectively. The network parameters are obtained by pre-training the network through the collected data based on the loss function.

6. The method for measuring vascular wall thickness based on coronary artery reconstruction according to claim 1, characterized in that: In S4, the specific method for measuring vascular wall thickness using the maximum inscribed sphere method is: The maximum endosphere method was used to make the centerline of the coronary artery and lumen. tol and l q , calculate the thickness of the blood vessel wall at each position, define the same plane perpendicular to the entrance of this section of blood vessels and the two midlines l tol and l q The intersection points are o tol and q The radii of the inscribed planes are r tol and r q , with o tol and q The direction of the line is the x-axis, o q Take O as the origin and make a plane coordinate system. The thickness of the blood vessel wall at any point on the plane is m i for: where θ i is the angle between the line from the inscribed plane i to the origin and the x-axis, and r is the midline l of the two planes tol and l q Intersection o tol and q The straight-line distance.

Images