Stent identification method and apparatus, and stent reconstruction method and apparatus

Abstract

The present application relates to the technical field of image processing. Provided are a stent identification method and apparatus, and a stent reconstruction method and apparatus. The stent identification method comprises: acquiring an original intravascular ultrasound (IVUS) image; performing anisotropic diffusion filtering on the original IVUS image to obtain an IVUS image in which stent signals are enhanced; processing the IVUS image in which the stent signals are enhanced, so as to form a binarized mask image that only includes the stent signals; and obtaining a stent identification result for the original IVUS image on the basis of the binarized mask image and the original IVUS image. Since IVUS has a greater imaging depth, even when the depth of a stent in a blood vessel is increased, stent signals can still be detected by means of IVUS, so that a situation where stents cannot be identified does not occur.

Description

A stent identification and reconstruction method and device Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to a method and device for stent recognition and reconstruction. Background Art

[0002] Intraluminal coronary stent placement refers to the placement of a metal mesh tube within a narrowed artery during percutaneous coronary intervention (PCI). This technique is used to support the vessel wall after balloon angioplasty, restore blood flow, and prevent acute vascular occlusion. After intraluminal coronary stent placement, the stent may not fully expand (the stent adheres correctly to the vessel wall but does not fully expand) or be misplaced (the stent does not fully contact the vessel wall). Incomplete stent deployment due to insufficient expansion and misplacement can lead to restenosis and thrombosis. Therefore, stent identification is necessary after stent placement to identify the stent's position and shape.

[0003] Traditional stent identification methods rely on intravascular optical coherence tomography (OCT). When a stent is first implanted, it is close to the vessel wall, allowing OCT to identify the stent. However, if restenosis occurs some time after implantation, the previously implanted stent may be inserted deeper within the vessel. Because OCT uses optical imaging, the imaging depth is shallow, making it difficult to identify the stent.

[0004] Summary of the Invention

[0005] In view of this, the present application provides a stent identification and reconstruction method and device to solve the problem of shallow OCT imaging depth in the prior art that stents cannot be identified. The technical solution is as follows:

[0006] A stent identification method, comprising:

[0007] Acquiring an original intravascular ultrasound (IVUS) image, wherein the original IVUS image is obtained by performing IVUS imaging on a stent in a blood vessel;

[0008] Perform anisotropic diffusion filtering on the original IVUS image to obtain an IVUS image with enhanced stent signals;

[0009] The IVUS image with enhanced stent signals was processed into a binary mask image containing only the stent signals;

[0010] The stent recognition result in the original IVUS image is obtained based on the binary mask image and the original IVUS image.

[0011] Optionally, anisotropic diffusion filtering is performed on the original IVUS image to obtain an IVUS image with enhanced stent signals, including:

[0012] Perform anisotropic diffusion filtering on the original IVUS image for a preset number of iterations, and determine the filtered image at the last iteration as the IVUS image with stent signal enhancement;

[0013] Among them, the process of anisotropic diffusion filtering under one iteration includes:

[0014] Calculate the four-directional nearest neighbor differences of the target image, calculate the four-directional conduction coefficients at this iteration based on the four-directional nearest neighbor differences of the target image, and calculate the pixel values ​​of the filtered image at this iteration based on the four-directional nearest neighbor differences of the target image and the four-directional conduction coefficients at this iteration to obtain the filtered image at this iteration, wherein, when this iteration is the first iteration, the target image is the original IVUS image, and when this iteration is not the first iteration, the target image is the filtered image at the previous iteration.

[0015] Optionally, the four-direction nearest neighbor differences include an east-direction nearest neighbor difference, a west-direction nearest neighbor difference, a south-direction nearest neighbor difference, and a north-direction nearest neighbor difference;

[0016] Calculate the four-way nearest neighbor difference of the target image, including:

[0017] For each pixel in the target image, calculate the difference between the pixel in the previous row and the first pixel of the pixel, and determine the calculated first pixel difference as the northern nearest neighbor difference of the pixel. If the pixel is in the first row of pixels in the target image, the pixel value of the pixel in the previous row is 0.

[0018] For each pixel in the target image, calculate the second pixel difference between the pixel and the corresponding pixel in the next row, and determine the calculated second pixel difference as the pixel's nearest neighbor difference in the south direction. If the pixel is in the last row of pixels in the target image, the pixel value of the corresponding pixel in the next row is 0.

[0019] For each pixel in the target image, calculate the third pixel difference between the pixel and the corresponding pixel in the previous column, and determine the calculated third pixel difference as the pixel's nearest neighbor difference in the west direction. If the pixel is in the first column of the target image, the pixel value of the corresponding pixel in the previous column is 0.

[0020] For each pixel in the target image, calculate the fourth pixel difference between the pixel's corresponding pixel in the next column and the pixel, and determine the calculated fourth pixel difference as the pixel's nearest neighbor difference in the east direction. If the pixel is the last column of pixels in the target image, the pixel value of the pixel's corresponding pixel in the next column is 0.

[0021] Optionally, the four-directional conduction coefficients include an east-direction conduction coefficient, a west-direction conduction coefficient, a south-direction conduction coefficient, and a north-direction conduction coefficient;

[0022] The four-directional conduction coefficients under this iteration are calculated based on the four-directional nearest neighbor differences of the target image, including:

[0023] Calculate the east direction conduction coefficient of this iteration according to the east direction nearest neighbor difference of the target image and the preset edge sensitivity;

[0024] Calculate the west-direction conduction coefficient under this iteration according to the west-direction nearest neighbor difference and edge sensitivity of the target image;

[0025] Calculate the south direction conduction coefficient under this iteration according to the south direction nearest neighbor difference and edge sensitivity of the target image;

[0026] The north direction conduction coefficient of this iteration is calculated based on the north direction nearest neighbor difference and edge sensitivity of the target image.

[0027] Optionally, the IVUS image with enhanced stent signals is processed into a binary mask image containing only stent signals, including:

[0028] The IVUS image with enhanced stent signals is binarized using a preset threshold to obtain a binarized image;

[0029] According to the shape of the stent and the size of the connected domain, the connected domain is deleted from the binarized image to obtain a binary mask image containing only the stent signal.

[0030] Optionally, a stent recognition result in the original IVUS image is obtained based on the binarized mask image and the original IVUS image, including:

[0031] The binary mask image and the original IVUS image are subjected to dot multiplication to obtain the stent recognition result in the original IVUS image.

[0032] A stent reconstruction method, comprising:

[0033] Acquire multiple original IVUS images, and use any of the above stent recognition methods to obtain stent recognition results in the multiple original IVUS images;

[0034] Obtain coronary angiography DSA images at two different angiography angles, and perform three-dimensional reconstruction of the blood vessel where the stent is located based on the two DSA images to obtain the blood vessel in three-dimensional space;

[0035] The stent recognition results in multiple original IVUS images are positioned and fused on the blood vessels in three-dimensional space to obtain the three-dimensional structure of the stent.

[0036] Optionally, the multiple original IVUS images are images acquired at equal intervals;

[0037] The stent recognition results from multiple original IVUS images are positioned and fused on the blood vessels in three-dimensional space to obtain the three-dimensional structure of the stent, including:

[0038] The stent recognition results in multiple original IVUS images are arranged at equal intervals in the blood vessels in the three-dimensional space to obtain the three-dimensional structure of the stent.

[0039] A bracket identification device, comprising:

[0040] An IVUS image acquisition module is used to acquire an original intravascular ultrasound IVUS image, wherein the original IVUS image is obtained by performing IVUS imaging on the stent in the blood vessel;

[0041] A stent signal enhancement module is used to perform anisotropic diffusion filtering on the original IVUS image to obtain an IVUS image with enhanced stent signals;

[0042] a mask image determination module, configured to process the IVUS image with enhanced stent signals into a binary mask image containing only stent signals;

[0043] The first stent recognition module is used to obtain a stent recognition result in the original IVUS image according to the binary mask image and the original IVUS image.

[0044] A stent reconstruction device, comprising:

[0045] A second stent recognition module is used to obtain multiple original IVUS images and obtain stent recognition results in the multiple original IVUS images using any of the above stent recognition methods;

[0046] The vascular 3D reconstruction module is used to obtain coronary angiography DSA images at two different angiography angles, and perform 3D reconstruction of the blood vessel where the stent is located based on the two DSA images to obtain the blood vessel in 3D space;

[0047] The stent 3D reconstruction module is used to locate and fuse the stent identification results in multiple original IVUS images on the blood vessels in three-dimensional space to obtain the three-dimensional structure of the stent.

[0048] As can be seen from the above technical solution, the stent identification method provided in this application uses intravascular ultrasound technology to obtain a raw IVUS image, then performs anisotropic diffusion filtering on the raw IVUS image to enhance the stent signal in the image, obtaining a stent-signal-enhanced IVUS image. The stent-signal-enhanced IVUS image is then processed into a binary mask image containing only the stent signal. Finally, based on the binary mask image and the original IVUS image, a stent identification result is obtained in the raw IVUS image. Because intravascular ultrasound has a deeper imaging depth, even if the stent is deeper within the blood vessel, the stent signal can still be detected by intravascular ultrasound, thus preventing the stent from being unidentified. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without any creative work.

[0050] FIG1 is a schematic diagram of a process of a stent identification method provided in an embodiment of the present application;

[0051] FIG2 is a schematic diagram of the positional relationship between pixels and the four-directional transmission coefficients provided in an embodiment of the present application;

[0052] FIG3( a ) is a schematic diagram of an original IVUS image with a stent provided in an embodiment of the present application;

[0053] FIG3( b ) is a schematic diagram of an IVUS image of stent signal enhancement provided in an embodiment of the present application;

[0054] FIG3( c ) is a two-dimensional intensity map of the original IVUS image provided in an embodiment of the present application;

[0055] FIG3( d ) is a two-dimensional intensity diagram of an IVUS image with enhanced stent signal provided in an embodiment of the present application;

[0056] FIG4 is a schematic flow chart of a stent reconstruction method provided in an embodiment of the present application;

[0057] FIG5 is a schematic diagram of discrete point coordinate mapping provided by an embodiment of the present application;

[0058] FIG6 is a schematic diagram of three-dimensional angiographic mapping provided in an embodiment of the present application;

[0059] FIG7 is a schematic diagram of stents arranged equidistantly in a three-dimensional blood vessel according to an embodiment of the present application;

[0060] FIG8 is a schematic structural diagram of a stent identification device provided in an embodiment of the present application;

[0061] FIG9 is a schematic structural diagram of a stent reconstruction device provided in an embodiment of the present application;

[0062] FIG10 is a hardware structure block diagram of a bracket identification device provided in an embodiment of the present application;

[0063] FIG11 is a hardware structure block diagram of a stent reconstruction device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0065] In order to facilitate the description of the stent identification, reconstruction method and device provided in this application, the relevant terms used in this application are now explained.

[0066] Intravascular ultrasound (IVUS) is a medical imaging method that combines non-invasive ultrasound technology with invasive catheter technology to perform cross-sectional morphological imaging of the cardiovascular system using a special catheter with an ultrasound probe connected to the end. As the gold standard for coronary artery examination, IVUS can visualize the internal morphology of blood vessels through cross-sectional morphological imaging.

[0067] Digital subtraction angiography (DSA) is a technology that visualizes blood vessel images by digitally processing angiographic images. DSA can image the overall morphology of blood vessels but cannot visualize the morphology inside blood vessels.

[0068] The present application provides a stent identification method, which will be described in detail through the following embodiments.

[0069] Please refer to FIG1 , which shows a schematic flow chart of a stent identification method provided in an embodiment of the present application. The stent identification method may include:

[0070] Step S101: Acquire an original intravascular ultrasound (IVUS) image.

[0071] The original IVUS image is obtained by performing IVUS imaging on the stent in the blood vessel.

[0072] Given that IVUS has a deeper penetration depth than OCT (1.5 mm for OCT and 5 mm for IVUS), in order to be able to identify the position of the stent after the stent is wrapped by vascular tissue (i.e., restenosis), IVUS imaging of the stent in the blood vessel can be performed using intravascular ultrasound technology to obtain the original IVUS image.

[0073] Because the stent is closer to the inner wall of the blood vessel during early implantation, and the acoustic impedance of the stent is significantly greater than that of the vascular tissue, the stent appears as a strong speckle signal on the original IVUS image. The present application can determine the stent position by the location of the special speckle formed by the stent struts on the original IVUS image.

[0074] Step S102: performing anisotropic diffusion filtering on the original IVUS image to obtain an IVUS image with enhanced stent signals.

[0075] This application can enhance the signal at the location of the stent through anisotropic diffusion filtering. Here, anisotropic diffusion filtering treats the image as a physical force field or thermal flow field, where image pixels always flow or move toward areas with similar values, so that areas with large differences (such as edges) are preserved.

[0076] Step S103: Process the IVUS image with enhanced stent signals into a binary mask image containing only stent signals.

[0077] Since the pixel difference between the stent position and other positions in the stent signal-enhanced IVUS image is large, the stent signal-enhanced IVUS image can be processed into a binary mask image containing only the stent signal by selecting an appropriate threshold.

[0078] In the present application, the pixel value at the support position in the binary mask image is different from the pixel values ​​at other positions. For example, optionally, the pixel value at the support position is 1, and the pixel values ​​at other positions are 0.

[0079] Step S104 : Obtain a stent recognition result in the original IVUS image based on the binary mask image and the original IVUS image.

[0080] The present application can mask the pixel values ​​at other positions in the original IVUS image through a binary mask image, leaving only the pixels at the stent position, and obtain the stent recognition result in the original IVUS image. The present application can determine the position of the stent in the blood vessel through the stent recognition result.

[0081] The stent identification method provided in this application uses intravascular ultrasound technology to obtain a raw IVUS image. The raw IVUS image is then subjected to anisotropic diffusion filtering to enhance the stent signal within the image, resulting in an IVUS image with enhanced stent signals. The enhanced IVUS image is then processed into a binary mask image containing only the stent signal. Finally, based on the binary mask image and the original IVUS image, the stent identification result in the raw IVUS image is obtained. Because IVUS imaging has a deeper depth, even if the stent is deeper within the vessel, IVUS can still detect the stent signal, thus preventing the stent from being unidentified.

[0082] The following describes "step S102, performing anisotropic diffusion filtering on the original IVUS image to obtain an IVUS image with enhanced stent signals" in the above embodiment.

[0083] In this embodiment, in order to better highlight the stent signal on the IVUS image, the original IVUS image can be subjected to multiple anisotropic diffusion filtering, that is, the present application can perform anisotropic diffusion filtering on the original IVUS image for a preset number of iterations, and determine the filtered image under the last iteration as the IVUS image with enhanced stent signal.

[0084] Optionally, the anisotropic diffusion filtering process in one iteration may include: calculating the four-directional nearest neighbor differences of the target image, calculating the four-directional conduction coefficients in the iteration based on the four-directional nearest neighbor differences of the target image, and calculating the pixel values ​​of the filtered image in the iteration based on the four-directional nearest neighbor differences of the target image and the four-directional conduction coefficients in the iteration, so as to obtain the filtered image in the iteration.

[0085] Wherein, when this iteration is the first iteration, the target image is the original IVUS image; when this iteration is not the first iteration, the target image is the filtered image in the previous iteration.

[0086] Specifically, the grayscale diffusion equation of the filtered image at the tth iteration is:

[0087] Where, I t represents the filtered image at the tth iteration, div represents the divergence operator, represents the gradient operator, △ represents the Laplace operator, and c(x, y, t) represents the conductivity coefficient (also known as the diffusion coefficient).

[0088] In this application, c(x, y, t) is a constant. When c(x, y, t) is a constant, the filtered image I under the tth iteration tThe grayscale diffusion equation can be equivalent to the isotropic heat diffusion equation shown in the following formula (2). t =cΔI Formula (2)

[0089] Suppose that at the tth iteration, we want to smooth a certain area of ​​the filtered image in the previous iteration, but we don’t want to smooth the boundaries of this area. We can set the thermal diffusivity to 1 inside each area and to 0 at the boundary. The blurring of the filtered image in the previous iteration will be completed within the area, but the boundary of the area will remain sharp.

[0090] In short, the anisotropic diffusion filter equates the original IVUS image to a thermal field, with image pixels acting as the heat diffused within the thermal field. When performing anisotropic diffusion filtering, the pixel change depends on the relationship between the pixel and the surrounding pixels within the current region. According to the thermal diffusivity coefficient set above, when there are no edge pixels around the pixel, its thermal diffusivity will be larger, and the diffused area will become smoother; if there are edge pixels around the pixel, its thermal diffusivity will be smaller, and diffusion will be weakened or even eliminated. By performing multiple anisotropic diffusion filters on the original IVUS image with the stent, tissue noise can be smoothed while preserving the stent.

[0091] Next, we introduce the process of performing anisotropic diffusion filtering on the target image.

[0092] The anisotropic diffusion, scale space, and edge detection of the target image can be described using a single formula. Formula (1) can be discretized on a square grid. The square grid is equivalent to the matrix of the target image. The pixels of the target image are at the vertices of the grid, and the conduction coefficients between pixels are associated with the connecting lines between the vertices of the grid. Taking the anisotropic diffusion filter at the t+1th iteration as an example, as shown in Figure 2, the four-nearest neighbor discretization of the Laplacian operator can be used to organize Formula (2) into the following Formula (3):

[0093] In the formula, λ represents the weight coefficient (0≤λ≤1 / 4), c N 、c S 、c E and c W represents the four-directional conduction coefficient (the four-directional conduction coefficient is the four components of the constant term c in the above formula (2)), and represents the nearest neighbor difference in four directions (the nearest neighbor difference is also the gradient operator mentioned above), N, S, E, and W represent the four directions of north, south, east, and west respectively. represents the pixel value at position (i, j) in the filtered image (in formula (3), the image is the target image) at the t-th iteration, Represents the pixel value at position (i, j) in the filtered image at the t+1th iteration.

[0094] From the above formula (3), we can know that the four-direction nearest neighbor differences in this application include: the nearest neighbor difference in the east direction West-direction nearest neighbor difference Nearest neighbor difference in the south direction Difference from the nearest neighbor in the north direction The four-directional conduction coefficients include: the east direction conduction coefficient c E , west-direction conduction coefficient c W , south-direction conduction coefficient c S and the north-direction conduction coefficient c N As long as the four-directional nearest neighbor differences and four-directional conduction coefficients are obtained, they can be substituted into formula (3) to obtain the pixel value of each pixel in the filtered image at the t+1th iteration, thereby obtaining the filtered image at the t+1th iteration.

[0095] Optionally, for each pixel in the target image, i，j (equivalent to the formula (3) For the convenience of subsequent description, the “t” is omitted) as an example, and the pixel I is calculated. i，j The process of the four-way nearest neighbor difference includes the following steps S11 to S14:

[0096] Step S11, calculate the difference between the pixel corresponding to the previous row of the pixel and the first pixel of the pixel, and determine the calculated first pixel difference as the nearest neighbor difference in the north direction of the pixel, wherein, if the pixel is the first row of pixels in the target image, the pixel value of the pixel corresponding to the previous row of the pixel is 0.

[0097] Specifically, the pixel I can be calculated using formula (4): i，j The difference in the nearest neighbor in the north direction

[0098] Where, I i-1，j Represents pixel I i，j The previous row corresponds to the pixel.

[0099] Step S12: Calculate the second pixel difference between the pixel and the pixel corresponding to the next row of pixels, and determine the calculated second pixel difference as the nearest neighbor difference of the pixel in the south direction. If the pixel is the last row of pixels in the target image, the pixel value of the pixel corresponding to the next row of pixels is 0.

[0100] Specifically, the pixel I can be calculated using formula (5) i，j The nearest neighbor difference in the south direction

[0101] Where, I i+1，j Represents pixel I i，j The next row corresponds to pixels.

[0102] Step S13: Calculate the third pixel difference between the pixel corresponding to the previous column and the pixel, and determine the calculated third pixel difference as the nearest neighbor difference of the pixel in the west direction. If the pixel is the first column of pixels in the target image, the pixel value of the pixel corresponding to the previous column is 0.

[0103] Specifically, the pixel I can be calculated using formula (6): i，j The west-direction nearest neighbor difference

[0104] Where, I i，j-1 Represents pixel I i，j The previous column corresponds to the pixel.

[0105] Step S14: Calculate the fourth pixel difference between the pixel corresponding to the next column of pixels and the pixel, and determine the calculated fourth pixel difference as the nearest neighbor difference of the pixel in the east direction. If the pixel is the last column of pixels in the target image, the pixel value of the pixel corresponding to the next column of pixels is 0.

[0106] Specifically, the pixel I can be calculated using formula (7) i，j The nearest neighbor difference in the east direction

[0107] Where, I i，j+1 Represents pixel I i，j The last column of corresponds to pixels.

[0108] In an optional embodiment, the process of “calculating the four-directional conduction coefficients in the iteration according to the four-directional nearest neighbor differences of the target image” is introduced.

[0109] It should be noted that although the four-directional transmission coefficients of each pixel do not change during one iteration, they are updated for different iterations. Therefore, the updated four-directional transmission coefficients need to be calculated at each iteration.

[0110] Optionally, the four-directional conduction coefficient can be calculated using the following formula (8):

[0111] Wherein, the constant term K is used to control the sensitivity to the edge and is usually selected based on the strength of the stent and the noise of the tissue.

[0112] As mentioned above, c has four components, namely c N 、c S 、c E and c W The above formula (8) can be rewritten into the following four components:

[0113] Where, Represents pixel I i，j The north-direction conduction coefficient, Represents pixel I i，j The southward transmission coefficient, Represents pixel I i，j The westward transmission coefficient, Represents pixel I i，j The eastward transmission coefficient.

[0114] According to the above formula (9), the present application can calculate the nearest neighbor difference in the north direction of the target image and edge sensitivity K, calculate the north direction conduction coefficient c under this iteration N , that is, for each pixel in the target image (with I i，j For example), according to the difference of the nearest neighbor in the north direction of the pixel and K, calculate the north direction conductivity of the pixel under this iteration

[0115] Accordingly, according to the above formula (10), the present application can calculate the nearest neighbor difference in the south direction of the target image. and edge sensitivity K, calculate the south direction conduction coefficient c under this iteration S , that is, for each pixel in the target image (with I i，j For example), based on the difference in the nearest neighbor in the south direction of the pixel and K, calculate the south direction conductivity coefficient of the pixel under this iteration

[0116] According to the above formula (11), the present application can calculate the nearest neighbor difference in the west direction of the target image and edge sensitivity K, calculate the west-direction conduction coefficient c under this iteration W , that is, for each pixel in the target image (with I i，j For example), according to the difference between the nearest neighbors in the west direction of the pixel and K, calculate the west-direction conduction coefficient of the pixel under this iteration

[0117] According to the above formula (12), the present application can calculate the nearest neighbor difference in the east direction of the target image And the preset edge sensitivity K, calculate the east direction transmission coefficient c under this iteration E , that is, for each pixel in the target image (with I i，j For example), according to the nearest neighbor difference in the east direction of the pixel and K, calculate the east direction conduction coefficient of the pixel under this iteration

[0118] In this application, the results obtained by the above formulas (4) to (7) and (9) to (12) are substituted into formula (3), and the result after anisotropic diffusion of the original IVUS image can be obtained, in which the stent signal is enhanced, while the tissue and noise signals are suppressed.

[0119] Taking the original IVUS image with a stent shown in Figure 3(a) as an example, the IVUS image with stent signal enhancement obtained after multiple anisotropic diffusion filtering is shown in Figure 3(b). Compared with Figure 3(a), the stent signal in Figure 3(b) is more obvious. In order to see the difference between the two images more clearly, the present application also provides two-dimensional intensity maps of the two images, wherein Figure 3(c) is a two-dimensional intensity map of the original IVUS image shown in Figure 3(a), and Figure 3(d) is a two-dimensional intensity map of the IVUS image with stent signal enhancement shown in Figure 3(b). Compared with Figure 3(c), Figure 3 (d) filters the noise pixels outside the stent signal very well, thereby highlighting the stent signal very well.

[0120] In summary, this application takes into account that the stent is close to the inner wall of the blood vessel during early implantation, and the acoustic impedance of the stent is significantly greater than the acoustic impedance of the vascular tissue. Therefore, the stent appears on the original IVUS image as speckle with a strong signal. The signal at the location of the stent is enhanced by anisotropic diffusion filtering, making it easier to distinguish the position of the stent in the IVUS image with enhanced stent signal.

[0121] An embodiment of the present application introduces the process of the aforementioned "step S103, processing the IVUS image with enhanced stent signals into a binary mask image containing only stent signals" and "step S104, obtaining the stent recognition result in the original IVUS image based on the binary mask image and the original IVUS image."

[0122] Specifically, the process of "step S103, processing the IVUS image with enhanced stent signals into a binary mask image containing only stent signals" may include:

[0123] Step S21 : binarizing the IVUS image with enhanced stent signals using a preset threshold to obtain a binarized image.

[0124] Here, all the scaffold signals are contained in the binarized image.

[0125] Specifically, because the present application pre-enhances the signals of the pixels representing the stent in the original IVUS image using anisotropic diffusion filtering, the stent-signal-enhanced IVUS image has larger pixel values ​​for those pixels representing the stent, while the pixel values ​​of the remaining pixels outside the stent are relatively small. Therefore, by setting an appropriate threshold, the pixels representing the stent can be easily distinguished from the remaining pixels, resulting in a binarized image.

[0126] For example, optionally, the binarization process of the present application includes: adjusting those pixel values ​​greater than or equal to a preset threshold in the IVUS image with enhanced stent signals to 1, and adjusting those pixel values ​​less than the preset threshold to 0, to obtain a binarized image.

[0127] Step S22: Deleting the connected domain of the binarized image according to the shape of the stent and the size of the connected domain to obtain a binarized mask image containing only the stent signal.

[0128] Specifically, this step can remove noise by removing smaller connected domains in the binarized image, thereby obtaining a binarized mask image containing only the support signal.

[0129] The method of removing connected domains is an existing technology and will not be described in detail here.

[0130] In this application, the IVUS image with enhanced stent signal is denoted as I S , to I S After binarization, the binarized image I can be obtained. BS , by removing I BS The smaller connected domain in the image is used to remove noise, and finally a binary mask image I of the bracket is obtained. Mask .

[0131] Because I Mask Only the bracket signal is included in the binary mask image I Mask Perform dot multiplication with the original IVUS image I to obtain the stent recognition result I in the original IVUS image Stent That is, in step S104, formula (13) can be used to obtain the stent recognition result I in the original IVUS image. Stent I Stent =I.*I MaskFormula (13)

[0132] Where, “.*” is the dot product symbol.

[0133] After the stent identification method provided in the aforementioned embodiment completes the segmentation of the stent, the three-dimensional reconstruction of the stent can also be achieved according to the embodiment of the present application, so that relevant personnel can timely understand the three-dimensional shape of the stent in the blood vessel.

[0134] To this end, an embodiment of the present application further provides a stent reconstruction method. See FIG4 , which is a flow chart of the stent reconstruction method provided in an embodiment of the present application.

[0135] As shown in FIG4 , the stent reconstruction method provided in this application may include:

[0136] Step S401: Acquire multiple original IVUS images, and use a stent recognition method to obtain stent recognition results in the multiple original IVUS images.

[0137] Specifically, IVUS can only visualize the internal morphology of the blood vessel through cross-sectional morphological imaging. However, the stent recognition result of an original IVUS image only reflects a cross-sectional morphology of the stent. If three-dimensional reconstruction of the stent is required, multiple original IVUS images need to be obtained, and then the stent recognition results in multiple original IVUS images are obtained using the stent recognition method provided in any of the above embodiments.

[0138] Step S402: Obtain coronary angiography DSA images at two different angiography angles, and perform three-dimensional reconstruction of the blood vessel where the stent is located based on the two DSA images to obtain the blood vessel in three-dimensional space.

[0139] This step is a process of obtaining blood vessels in three-dimensional space based on DSA images at two different angiography angles. This is a prior art process. To make this application more complete, a brief introduction is given below.

[0140] Specifically, for the three-dimensional reconstruction of vascular structures, it can be converted into a problem of three-dimensional reconstruction of discrete points under two different angiography angles. The principle of three-dimensional reconstruction is explained in conjunction with Figure 5 below. Assume that S1 and S2 are ray sources, U1V1O1 and U2V2O2 are two acquisition planes of angiography (i.e., DSA images under two different angiography angles), S1O1 and S2O2 are axes from the centers of the two acquisition planes to the centers of the projection coordinate systems X1Y1Z1S1 and X2Y2Z2S2, α1 and α2 represent the left and right angles of the two angiography images relative to the coordinate axis XYZO, β1 and β2 represent the front and back angles of the angiography, where α1 refers to the angle formed by A1OZ, α2 refers to the angle formed by A2OZ, β1 refers to the angle formed by A2OO2, and β2 refers to the angle formed by A1OO1. Take a point on each of the two acquisition planes, for example, take p1j (u 1j , v 1j ) and p 2j (u 2j , v 2j ), assuming p 1j and p 2j These are two points on the same blood vessel at the same position on the angiogram. The three-dimensional reconstruction of blood vessels can be explained by taking two points as an example. i Y i Z i S i (i.e. X1Y1Z1S1 and X2Y2Z2S2 above), p 1j and p 2j The coordinates in it are (x i ,y i , z i (Specifically, in X1Y1Z1S1 it is (x1, y1, z1), in X2Y2Z2S2 it is (x2, y2, z2). According to the geometric relationship, the geometric relationship of the coordinate system X1Y1Z1S1 transformed to X2Y2Z2S2 is:

[0141] Where R is the rotation matrix, For the translation matrix, the coordinate system XYZO to X can also be realized by translation and rotation i Y i Z i S i The transformation of [x i ,y i , z i ,1]=[X,Y,Z,1]·R Y (α i )·R X (β i )·T i Formula (15)

[0142] Among them, T i Represents translation along the Z axis, matrix R X (β i ) represents a clockwise rotation around the X axis β i Angle, R Y (α i ) represents a clockwise rotation around the Y axis α i Angle, where β i Including the above β1 and β2, α i Including α1 and α2. In actual situations, it is known that (u 1j , v 1j ) and (u 2j , v 2j), p can be solved according to the translation and rotation matrix j The three-dimensional coordinates of p j refers to p 1j S1 and p 2j The intersection of S2.

[0143] Similarly, by selecting several pairs of matching points on the two DSA images, the two-dimensional blood vessels can be mapped into the blood vessel structure in the three-dimensional space as shown in FIG6 .

[0144] Step S403: Positioning and fusing the stent recognition results in the multiple original IVUS images on the blood vessels in the three-dimensional space to obtain the three-dimensional structure of the stent.

[0145] In an optional embodiment, when acquiring multiple original IVUS images in step S401, multiple original IVUS images may be acquired at equal intervals. When the multiple original IVUS images are acquired at equal intervals, it can be assumed that the reconstructed vascular spatial structure based on the acquired images is also equally spaced. Therefore, step S403 may include arranging the stent identification results from the multiple original IVUS images at equal intervals within the blood vessel in three-dimensional space to obtain a three-dimensional structure of the stent, as shown in FIG7 .

[0146] In summary, the embodiment of the present application performs IVUS imaging of the imaging target in the early stage of stent implantation, and DSA imaging is used as an auxiliary in the process of IVUS imaging. Since the stent is closer to the inner wall of the blood vessel during early implantation and the acoustic impedance of the stent is significantly greater than the acoustic impedance of the vascular tissue, the stent is reflected in the original IVUS image as a speckle with a strong signal. The signal at the location of the stent is enhanced by anisotropic diffusion filtering, and the enhanced image is binarized by setting a measurement threshold. At this time, all the stent signals are included in the binary image, and then the binary image is filtered according to the shape of the stent and the size of its connected domain. Connected domains are deleted to obtain a binary image containing only the stent signal. This binary image is used as a mask, and the matrix of stent position and intensity is obtained by matrix multiplication of the stent signal-enhanced IVUS image. The point spread function (PSF) of the stent is measured and a normalized two-dimensional cross-correlation is performed with the matrix obtained in the previous step. An appropriate threshold is set to remove unnecessary noise, and the spatial coordinates of the centroid of the cross-correlated image are extracted one by one to locate the individual stent points. Finally, combined with the vascular spatial position and structural information obtained by DSA, the stent is accumulated frame by frame to the corresponding position of the vessel to obtain the final three-dimensional stent image.

[0147] This application provides a method for stent identification and positioning (three-dimensional reconstruction) using intravascular ultrasound images, which provides an effective method for stent implantation and immediate effect evaluation and analysis after stent implantation. Combined with DSA for three-dimensional reconstruction of the stent, it can visualize the structure and spatial position of the stent, providing more information in the early stages of stent implantation.

[0148] The embodiment of the present application further provides a bracket identification device, which is described below.

[0149] Please refer to Figure 8, which shows a structural schematic diagram of the stent identification device provided in an embodiment of the present application. As shown in Figure 8, the stent identification device may include: an IVUS image acquisition module 801, a stent signal enhancement module 802, a mask image determination module 803 and a first stent identification module 804.

[0150] The IVUS image acquisition module 801 is used to acquire an original intravascular ultrasound IVUS image, wherein the original IVUS image is obtained by performing IVUS imaging on the stent in the blood vessel.

[0151] The stent signal enhancement module 802 is used to perform anisotropic diffusion filtering on the original IVUS image to obtain an IVUS image with enhanced stent signals.

[0152] The mask image determination module 803 is configured to process the IVUS image with enhanced stent signals into a binary mask image containing only stent signals.

[0153] The first stent recognition module 804 is configured to obtain a stent recognition result in the original IVUS image according to the binary mask image and the original IVUS image.

[0154] The working principle of the stent identification device provided in this application is the same as the working principle of the stent identification method described above. For details, please refer to the above introduction and will not be repeated here.

[0155] The embodiment of the present application further provides a stent reconstruction device, which is described below.

[0156] Please refer to Figure 9, which shows a structural diagram of a stent reconstruction device provided in an embodiment of the present application. As shown in Figure 9, the stent reconstruction device may include: a second stent identification module 901, a blood vessel three-dimensional reconstruction module 902 and a stent three-dimensional reconstruction module 903.

[0157] The second stent recognition module 901 is used to acquire multiple original IVUS images and obtain stent recognition results in the multiple original IVUS images using a stent recognition method.

[0158] The blood vessel three-dimensional reconstruction module 902 is used to obtain coronary angiography DSA images at two different angiography angles, and perform three-dimensional reconstruction of the blood vessel where the stent is located based on the two DSA images to obtain the blood vessel in three-dimensional space.

[0159] The stent 3D reconstruction module 903 is used to position and fuse the stent recognition results in multiple original IVUS images on the blood vessels in the 3D space to obtain the 3D structure of the stent.

[0160] The working principle of the stent reconstruction device provided in this application is the same as the working principle of the stent reconstruction method described above. Please refer to the above introduction for details and will not be repeated here.

[0161] The embodiment of the present application also provides a bracket identification device. Optionally, FIG10 shows a hardware structure block diagram of the bracket identification device. Referring to FIG10 , the hardware structure of the bracket identification device may include: at least one processor 1001, at least one communication interface 1002, at least one memory 1003, and at least one communication bus 1004;

[0162] In the embodiment of the present application, the number of the processor 1001, the communication interface 1002, the memory 1003, and the communication bus 1004 is at least one, and the processor 1001, the communication interface 1002, and the memory 1003 communicate with each other through the communication bus 1004;

[0163] The processor 1001 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention;

[0164] The memory 1003 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory;

[0165] The memory 1003 stores a program, and the processor 1001 can call the program stored in the memory 1003, and the program is used to:

[0166] Acquiring an original intravascular ultrasound (IVUS) image, wherein the original IVUS image is obtained by performing IVUS imaging on a stent in a blood vessel;

[0167] Perform anisotropic diffusion filtering on the original IVUS image to obtain an IVUS image with enhanced stent signals;

[0168] The IVUS image with enhanced stent signals was processed into a binary mask image containing only the stent signals;

[0169] The stent recognition result in the original IVUS image is obtained based on the binary mask image and the original IVUS image.

[0170] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0171] The present application also provides a stent reconstruction device. Optionally, FIG11 shows a hardware structure block diagram of the stent reconstruction device. Referring to FIG11 , the hardware structure of the stent reconstruction device may include: at least one processor 1101, at least one communication interface 1102, at least one memory 1103, and at least one communication bus 1104.

[0172] In the embodiment of the present application, the number of the processor 1101, the communication interface 1102, the memory 1103, and the communication bus 1104 is at least one, and the processor 1101, the communication interface 1102, and the memory 1103 communicate with each other through the communication bus 1104;

[0173] The processor 1101 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention;

[0174] The memory 1103 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory;

[0175] The memory 1103 stores a program, and the processor 1101 can call the program stored in the memory 1103, and the program is used to:

[0176] Acquire multiple original IVUS images, and use a stent recognition method to obtain stent recognition results in the multiple original IVUS images;

[0177] Obtain coronary angiography DSA images at two different angiography angles, and perform three-dimensional reconstruction of the blood vessel where the stent is located based on the two DSA images to obtain the blood vessel in three-dimensional space;

[0178] The stent recognition results in multiple original IVUS images are positioned and fused on the blood vessels in three-dimensional space to obtain the three-dimensional structure of the stent.

[0179] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0180] An embodiment of the present application further provides a readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the above-mentioned bracket identification method is implemented.

[0181] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0182] An embodiment of the present application further provides a readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the stent reconstruction method as described above is implemented.

[0183] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0184] Finally, it should be noted that, in this document, relational terms such as and and the like are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0185] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0186] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying a stent, characterized in that: include: Acquire an original intravascular ultrasound (IVUS) image, wherein the original IVUS image is obtained by performing IVUS imaging on a stent in a blood vessel; Performing anisotropic diffusion filtering on the original IVUS image to obtain an IVUS image with enhanced stent signals; Processing the stent signal-enhanced IVUS image into a binary mask image containing only the stent signal; The stent recognition result in the original IVUS image is obtained according to the binary mask image and the original IVUS image.

2. The stent identification method according to claim 1, characterized in that: The step of performing anisotropic diffusion filtering on the original IVUS image to obtain an IVUS image with enhanced stent signals includes: Performing anisotropic diffusion filtering for a preset number of iterations on the original IVUS image, and determining the filtered image at the last iteration as the IVUS image with enhanced stent signals; Among them, the process of anisotropic diffusion filtering under one iteration includes: Calculate the four-directional nearest neighbor differences of the target image, calculate the four-directional conduction coefficients at this iteration based on the four-directional nearest neighbor differences of the target image, and calculate the pixel values ​​of the filtered image at this iteration based on the four-directional nearest neighbor differences of the target image and the four-directional conduction coefficients at this iteration to obtain the filtered image at this iteration, wherein, when this iteration is the first iteration, the target image is the original IVUS image, and when this iteration is not the first iteration, the target image is the filtered image at the previous iteration.

3. The stent identification method according to claim 2, characterized in that: The four-direction nearest neighbor differences include the nearest neighbor difference in the east direction, the nearest neighbor difference in the west direction, the nearest neighbor difference in the south direction, and the nearest neighbor difference in the north direction; The calculating of the nearest neighbor differences of the target image in four directions includes: For each pixel in the target image, a difference between a pixel corresponding to the previous row of the pixel and the first pixel of the pixel is calculated, and the calculated first pixel difference is determined as the nearest neighbor difference in the north direction of the pixel, wherein if the pixel is a pixel in the first row of the target image, the pixel value of the pixel corresponding to the previous row of the pixel is 0; For each pixel in the target image, calculate the second pixel difference between the pixel and the pixel in the next row, and determine the calculated second pixel difference as the nearest neighbor difference in the south direction of the pixel, wherein if the pixel is the last row of pixels in the target image, the pixel value of the pixel in the next row of pixels is 0; For each pixel in the target image, calculate the third pixel difference between the pixel and the corresponding pixel in the previous column, and determine the calculated third pixel difference as the nearest neighbor difference in the west direction of the pixel, wherein if the pixel is a pixel in the first column of the target image, the pixel value of the corresponding pixel in the previous column of the pixel is 0; For each pixel in the target image, the difference between the pixel corresponding to the next column and the fourth pixel of the pixel is calculated, and the calculated fourth pixel difference is determined as the nearest neighbor difference of the pixel in the east direction, wherein, if the pixel is the last column of pixels in the target image, the pixel value of the pixel corresponding to the next column of the pixel is 0.

4. The method for identifying a stent according to claim 3, characterized in that: The four-directional conduction coefficients include an east-direction conduction coefficient, a west-direction conduction coefficient, a south-direction conduction coefficient and a north-direction conduction coefficient; The calculating the four-directional transmission coefficients in the iteration according to the four-directional nearest neighbor differences of the target image includes: Calculate the east direction conduction coefficient of the iteration according to the east direction nearest neighbor difference of the target image and the preset edge sensitivity; Calculating the west direction conduction coefficient under this iteration according to the west direction nearest neighbor difference of the target image and the edge sensitivity; Calculating the south direction conduction coefficient of the iteration according to the south direction nearest neighbor difference of the target image and the edge sensitivity; The north direction conduction coefficient in this iteration is calculated according to the north direction nearest neighbor difference of the target image and the edge sensitivity.

5. The stent identification method according to claim 1, characterized in that: The step of processing the IVUS image with enhanced stent signals into a binary mask image containing only stent signals comprises: Binarization is performed on the IVUS image with enhanced stent signal by using a preset threshold value to obtain a binarized image; According to the shape of the support and the size of the connected domain, the connected domain is deleted from the binarized image to obtain the binarized mask image containing only the support signal.

6. The stent identification method according to claim 1, characterized in that: Obtaining a stent recognition result in the original IVUS image according to the binary mask image and the original IVUS image, including: The binary mask image and the original IVUS image are subjected to a dot multiplication process to obtain a stent recognition result in the original IVUS image.

7. A stent reconstruction method, characterized in that: include: Acquire multiple original IVUS images, and use the stent recognition method described in any one of claims 1 to 4 to obtain stent recognition results in the multiple original IVUS images; Obtaining coronary angiography DSA images at two different angiography angles, and performing three-dimensional reconstruction of the blood vessel where the stent is located according to the two DSA images to obtain the blood vessel in the three-dimensional space; The stent recognition results in the multiple original IVUS images are positioned and fused on the blood vessels in the three-dimensional space to obtain the three-dimensional structure of the stent.

8. The stent reconstruction method according to claim 7, characterized in that: The multiple original IVUS images are images collected at equal intervals; The step of positioning and fusing the stent recognition results in the plurality of original IVUS images on the blood vessels in the three-dimensional space to obtain the three-dimensional structure of the stent includes: The stent recognition results in the multiple original IVUS images are arranged at equal intervals in the blood vessels in the three-dimensional space to obtain the three-dimensional structure of the stent.

9. A bracket identification device, characterized in that: include: An IVUS image acquisition module, used to acquire an original intravascular ultrasound IVUS image, wherein the original IVUS image is obtained by performing IVUS imaging on the stent in the blood vessel; A stent signal enhancement module, used for performing anisotropic diffusion filtering on the original IVUS image to obtain an IVUS image with enhanced stent signals; A mask image determination module, used for processing the IVUS image with enhanced stent signals into a binary mask image containing only stent signals; The first stent recognition module is used to obtain a stent recognition result in the original IVUS image according to the binary mask image and the original IVUS image.

10. A stent reconstruction device, characterized in that: include: A second stent recognition module, used to obtain a plurality of original IVUS images, and to obtain stent recognition results in the plurality of original IVUS images using the stent recognition method according to any one of claims 1 to 4; A blood vessel three-dimensional reconstruction module is used to obtain coronary angiography DSA images at two different angiography angles, and to perform three-dimensional reconstruction of the blood vessel where the stent is located based on the two DSA images to obtain the blood vessel in the three-dimensional space; The stent three-dimensional reconstruction module is used to locate and fuse the stent recognition results in the multiple original IVUS images on the blood vessels in the three-dimensional space to obtain the three-dimensional structure of the stent.

Images