Aortic dissection image acquisition method and device based on non-enhanced CT image

Abstract

The invention discloses an aortic dissection image acquisition method and device based on a non-enhanced CT image. The aortic dissection image acquisition method based on the non-enhanced CT image comprises the following steps: acquiring original image information; acquiring a preprocessing feature matrix according to the original image information; inputting the pre-processed feature matrix into a two-stage cascade calcification internal shift recognition network to obtain a calcification internal shift recognition image; inputting the preprocessing feature matrix into an inner diaphragm recognition network to obtain an inner diaphragm recognition image; and inputting the preprocessed feature matrix into a pseudo CTA image generation network to obtain a pseudo CTA image. According to the aortic dissection image acquisition method based on the non-enhanced CT image, the layer and the position of the dissection can be clearly prompted, and a doctor is assisted to quickly confirm whether the NCCT related layer contains the suspected dissection or not.

Description

Technical Field

[0001] The present application relates to the field of image recognition technology, and in particular to a method for acquiring an aortic dissection image based on a non-enhanced CT image and a device for acquiring an aortic dissection image based on a non-enhanced CT image. Background Art

[0002] Aortic dissection refers to the blood in the aorta cavity entering the aortic media from the tear in the aortic intima, causing the media to separate and expand along the long axis of the aorta to form a state of separation between the true and false cavities of the aortic wall. This disease is rare, with an incidence rate of 1 in 100,000 to 1 in 200,000 per year. The peak age is 50 to 70 years old, and the male to female ratio is about 2 to 3:1. 65% to 70% die from cardiac tamponade, arrhythmia, etc. in the acute stage, so early diagnosis and treatment are very necessary.

[0003] The diagnosis of aortic dissection mainly relies on imaging examinations. Commonly used screening methods in clinical practice include transthoracic echocardiography (TTE), CT angiography (CTA), and magnetic resonance imaging (MRI). CT angiography (CTA) is a noninvasive and effective vascular imaging technique that is widely used to diagnose vascular lesions such as aortic aneurysms and dissections. However, CTA requires intravascular injection of iodinated contrast agents (ICA), which poses risks to patients with iodine allergy and renal insufficiency. In addition, compared with plain CT (NCCT), CTA is expensive and time-consuming, especially in grassroots areas where technology and support are lacking, and the application of 24-hour CTA is limited.

[0004] It is very difficult for doctors with little experience to identify aortic dissection from plain scan CT, which can easily lead to missed diagnosis. Therefore, it is meaningful to develop a method to automatically detect high-risk signs of aortic dissection from plain scan CT (NCCT) images, which can reduce missed diagnosis of aortic dissection.

[0005] Therefore, it is hoped that a technical solution can be provided to solve or at least alleviate the above-mentioned deficiencies of the prior art. Summary of the invention

[0006] The object of the present invention is to provide a method for acquiring an aortic dissection image based on a non-enhanced CT image to at least solve one of the above-mentioned technical problems.

[0007] The present invention provides the following scheme:

[0008] According to one aspect of the present invention, a method for acquiring an aortic dissection image based on a non-enhanced CT image is provided, and the method for acquiring an aortic dissection image based on a non-enhanced CT image comprises:

[0009] Get original image information;

[0010] Preprocessing the original image information to obtain a preprocessing feature matrix;

[0011] Obtain a two-level cascade calcification inward migration recognition network, an intimal sheet recognition network, and a pseudo-CTA image generation network;

[0012] Inputting the preprocessed feature matrix into the two-stage cascade calcification inward shift recognition network, thereby obtaining a calcification inward shift recognition image;

[0013] Inputting the preprocessed feature matrix into the endothelial sheet recognition network to obtain an endothelial sheet recognition image;

[0014] The preprocessed feature matrix is ​​input into the pseudo CTA image generation network to obtain a pseudo CTA image.

[0015] Optionally, preprocessing the original data to obtain a preprocessing matrix includes:

[0016] The original data are subjected to windowing and voxel spacing normalization.

[0017] Optionally, the two-stage cascade neural network includes:

[0018] The first-level NCCT aorta segmentation network adopts the nnUnet architecture;

[0019] The second-level calcification inward migration segmentation network, the output of the first-level NCCT aorta segmentation network is used as the input of the second-level calcification inward migration segmentation network; wherein,

[0020] The downsampling step size of the z-direction convolution kernel of the network backbone layer of the two-stage cascade neural network is changed to 1.

[0021] Optionally, the endocardial sheet recognition network includes a 3D segmentation network and a 2D segmentation network.

[0022] Optionally, the pseudo CTA image generation network includes a generator, a discriminator and a registration network, the original image information is converted into a pseudo CTA image through the generator, and the original image information and the pseudo CTA image are respectively sent to the discriminator to obtain results.

[0023] The present application also provides an aortic dissection image acquisition device based on a non-enhanced CT image, characterized in that the aortic dissection image acquisition device based on a non-enhanced CT image comprises:

[0024] An original image information acquisition module, wherein the original image information acquisition module is used to acquire original image information;

[0025] A preprocessing feature matrix acquisition module, wherein the preprocessing feature matrix acquisition module is used to preprocess the original image information to obtain a preprocessing feature matrix;

[0026] A network acquisition module, the network acquisition module is used to acquire a two-stage cascade calcification inward migration recognition network, an intimal sheet recognition network, and a pseudo CTA image generation network;

[0027] A calcification inward shift recognition image acquisition module, the calcification inward shift recognition image acquisition module is used to input the preprocessing feature matrix into the two-stage cascade calcification inward shift recognition network, so as to acquire a calcification inward shift recognition image;

[0028] An endothelial patch recognition image acquisition module, wherein the endothelial patch recognition image acquisition module is used to input the preprocessed feature matrix into the endothelial patch recognition network, thereby acquiring an endothelial patch recognition image;

[0029] A pseudo CTA image acquisition module is used to input the preprocessed feature matrix into the pseudo CTA image generation network to acquire a pseudo CTA image.

[0030] The aortic dissection image acquisition method based on non-enhanced CT images of the present application can clearly indicate the layer and position of the dissection, and assist doctors to quickly confirm whether the relevant layer of NCCT contains suspected dissection. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a flowchart of a method for acquiring an aortic dissection image based on a non-enhanced CT image in one embodiment of the present application.

[0032] Figure 2 It is a schematic diagram of the structure of an electronic device in one embodiment of the present application.

[0033] Figure 3 It is a schematic diagram of an image of the aorta of a patient with positive dissection in the prior art.

[0034] Figure 4 This is a schematic diagram of the structure of a two-stage cascade calcification inward migration recognition network of the aortic dissection image acquisition method based on non-enhanced CT images of the present application.

[0035] Figure 5 This is a schematic diagram of the structure of the intimal patch recognition network of the aortic dissection image acquisition method based on non-enhanced CT images of the present application.

[0036] Figure 6 This is a schematic diagram of the structure of a pseudo CTA image generation network of the aortic dissection image acquisition method based on non-enhanced CT images of the present application. DETAILED DESCRIPTION

[0037] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] Figure 1 It is a flowchart of a method for acquiring an aortic dissection image based on a non-enhanced CT image in one embodiment of the present application.

[0039] like Figure 1 The aortic dissection image acquisition method based on non-enhanced CT images includes:

[0040] Step 1: Get the original image information;

[0041] Step 2: preprocessing the original image information to obtain a preprocessing feature matrix;

[0042] Step 3: Obtain a two-level cascade calcification inward migration recognition network, an intimal sheet recognition network, and a pseudo-CTA image generation network;

[0043] Step 4: inputting the preprocessed feature matrix into the two-stage cascade calcification inward shift recognition network to obtain a calcification inward shift recognition image;

[0044] Step 5: inputting the preprocessed feature matrix into the endothelial sheet recognition network to obtain an endothelial sheet recognition image;

[0045] Step 6: Input the preprocessed feature matrix into the pseudo CTA image generation network to obtain a pseudo CTA image.

[0046] The aortic dissection image acquisition method based on non-enhanced CT images of the present application can clearly indicate the layer and position of the dissection, and assist doctors to quickly confirm whether the relevant layer of NCCT contains suspected dissection.

[0047] In this embodiment, the preprocessing of the original data to obtain a preprocessing matrix includes:

[0048] The original data are subjected to windowing and voxel spacing normalization.

[0049] In this embodiment, the windowing operation refers to setting the Hu value range of the original plain scan CT image to a specific window width and window position (this patent uses a mediastinum window with a window width of 500 and a window position of 50) to enhance the contrast of the image within a specific grayscale range.

[0050] Voxel spacing normalization refers to scaling the original plain scan CT image to a uniform voxel spacing. This patent uses a pixel spacing of 1 mm (x, y directions) and a layer spacing of 5 mm (z direction) for thick-layer CT.

[0051] See also Figure 4 In this embodiment, the two-stage cascade neural network includes a first-stage NCCT aorta segmentation network and a second-stage calcification inward migration segmentation network, wherein:

[0052] The first-level NCCT aorta segmentation network adopts the nnUnet architecture;

[0053] The output of the first-level NCCT aorta segmentation network is used as the input of the second-level calcification inward migration segmentation network; wherein,

[0054] The downsampling step size of the z-direction convolution kernel of the backbone layer of the two-stage cascade neural network is changed to 1.

[0055] In this embodiment, based on NCCT, the present application performs large-scale voxel-level calcification inward migration annotation and trains a 3D segmentation network. The network architecture is as follows: Figure 4 As shown in the figure, the recognition of aortic calcification inward migration is realized by using a segmentation network based on deep learning. It is divided into two levels of cascade network, the first level is the NCCT aortic segmentation network, and the second level is the calcification inward migration segmentation network. In the cascade network architecture, the output aortic contour of the first level provides the second level network with supporting judgment information of calcification inward migration, which can effectively suppress the false positive of calcification inward migration signs.

[0056] The calcification inward migration segmentation network adopts a two-stage cascade structure, rather than a single stage directly following the calcification. Its advantage is that the output results of the first-stage NCCT aortic segmentation network can provide important aortic lumen distance information for the second-stage network, making it easy for the second-stage network to identify whether the calcification point is close to the aortic wall or biased towards the inside of the aortic lumen. On the contrary, the second-stage network needs to learn the morphological characteristics of the calcification point and the range constraints of the aortic lumen separately, which is not conducive to the detection of calcification inward migration.

[0057] The first-level network uses the nnUnet architecture, and the second-level network uses the ConvNeXt architecture. If the input image is a thick layer image, the following improvements are made to the network: the z-direction convolution kernel of the network stem layer is changed to 1 in the downsampling stride. In this way, the network stem layer only downsamples in the x and y axial method, but not in the z direction. This can fully retain the z-direction information and avoid the invisible calcification lesions caused by oversampling of thick layer images.

[0058] In this embodiment, the endocardial sheet recognition network includes a 3D segmentation network and a 2D segmentation network.

[0059] In this embodiment, the present application relies on paired NCCT and CTA data to annotate large-scale interlayer endocardial slices and train a 3D and 2D fusion segmentation network. The network architecture is as follows Figure 5 As shown in Figure 2, unlike calcification, it is difficult to directly identify the intimal sheet on NCCT images. Therefore, it is necessary to collect NCCT and CTA paired data of the same patient for intimal sheet annotation.

[0060] In order to distinguish true endothelial sheets from various artifacts and false positive signs caused by bone volume effects, the network needs to refer to long-distance image context information. Making full use of the image continuity of adjacent layers is conducive to accurately distinguishing true endothelial sheets from false positive endothelial sheets. In this case, the 3D network is more advantageous than the 2D network (axial slices). On the other hand, considering that thick-layer (about 5mm) plain scan CT is widely used in clinical practice, the inter-layer spacing is significantly larger than the pixel spacing of axial slices. In order to achieve accurate segmentation of the axial contour of the endothelial sheet, it is necessary to learn high-resolution axial layer image detail features. Considering the above two aspects, this patent adopts a 3D+2D fusion model to achieve endothelial sheet segmentation.

[0061] In this embodiment, the pseudo-CTA image generation network includes a generator and a discriminator. The real NCCT image is converted into a "pseudo-CTA" image through the "generator". The "real CTA" and "pseudo-CTA" are respectively sent to the discriminator. The "authenticity" of the pseudo-CTA image is improved through adversarial loss. Due to the relative displacement caused by the patient's breathing and other reasons, the paired CTA and NCCT cannot correspond pixel by pixel. In order to use the "pseudo" CTA generated by pixel-by-pixel supervision of the CTA image, a registration network is introduced. At the same time, the "registration loss" is used to constrain the registration transformation deformation field to make the deformation field as smooth as possible. After the "pseudo-CTA" image is registered to the real CTA, a pixel-by-pixel loss function can be used to supervise the training of the generator network and the registration network, such as the L1 and L2 loss functions.

[0062] Specifically, the discriminator of the present application consists of 4 levels of convolutional layers, 1 level of fully connected layer, and 1 pooling layer. Both the generator and the registration network are U-net networks, where the generator contains 2 encoding paths, 1 bottleneck layer, 2 decoding paths, and 1 output layer. The registration network is a ResUnet network structure, which contains 7 encoding paths, 1 bottleneck layer, 7 decoding paths, and 1 output layer.

[0063] In this embodiment, the U-Net network includes the following parts:

[0064] (1) Contracting Path: It consists of a series of convolutional layers and max pooling layers. Each convolutional block contains two 3x3 convolutional layers, each followed by a ReLU activation function. Each convolutional block is followed by a 2x2 max pooling layer for downsampling (usually with a stride of 2). The contracting path gradually reduces the spatial dimension of the feature map while increasing the number of channels of the feature map.

[0065] (2) Bottleneck Layer: Located at the bottom of the U-shaped network, it connects the encoding path and the decoding path. It contains two 3x3 convolutional layers and a ReLU activation function, and does not perform pooling operations.

[0066] The number of feature map channels here is further increased compared to the encoding path, and the features are richer.

[0067] (3) Decoding path (Expansive Path): Symmetric to the encoding path, it consists of a series of upsampling layers and convolutional layers. Each upsampling block first performs a 2x2 upsampling (deconvolution) operation to restore the spatial dimension. After upsampling, the feature map is concatenated (skip connection) with the corresponding encoding path feature map in the channel dimension. The concatenated feature map is processed by two 3x3 convolutional layers, each followed by a ReLU activation function.

[0068] (4) Output Layer: The last convolutional layer uses 1x1 convolution to reduce the number of channels to the required number of categories. The final output feature map has the same spatial dimension as the input image, and each pixel corresponds to a category probability. The present invention relies on paired NCCT and CTA data to design and train a pseudo-CTA image generation network. In real NCCT images, the imaging features of the dissection are usually not obvious. Pseudo-CTA highlights the imaging signs of aortic dissection visible on NCCT in a form similar to angiography.

[0069] The training process of the pseudo CTA image generation network is as follows: Figure 6As shown in the figure. The real NCCT image is converted into a "pseudo CTA" image through the "generator". Based on the idea of ​​adversarial generative network, the "real CTA" and "pseudo CTA" are sent to the discriminator respectively, and the "authenticity" of the pseudo CTA image is improved through adversarial loss. Due to the relative displacement caused by the patient's breathing and other reasons, the paired CTA and NCCT cannot correspond pixel by pixel. In order to use the "pseudo" CTA generated by pixel-by-pixel supervision of CTA images, a registration network is introduced. At the same time, the "registration loss" is used to constrain the registration transformation deformation field to make the deformation field as smooth as possible. After registering to the real CTA, the "pseudo CTA" image can use a pixel-by-pixel loss function to supervise the training of the generator network and the registration network, such as the L1 and L2 loss functions.

[0070] In the pseudo-CTA generation module, the recognition results of "NCCT calcification inward shift" and "NCCT endothelial sheet" are used as constraints of the generator, which has the following advantages: (1) Compared with unsupervised image conversion training, "NCCT calcification inward shift" and "NCCT endothelial sheet" are discriminant networks, and supervised training is used, so the reliability and interpretability of their detection signs are stronger. (2) The generator can combine the results of the above model with NCCT images, so that the generated images are more consistent with the morphology of real CTA.

[0071] The present application also provides an aortic dissection image acquisition device based on non-enhanced CT images, the aortic dissection image acquisition device based on non-enhanced CT images comprises an original image information acquisition module, a pre-processing feature matrix acquisition module, a network acquisition module, a calcification inward migration recognition image acquisition module, an intimal patch recognition image acquisition module and a pseudo CTA image acquisition module, wherein:

[0072] The original image information acquisition module is used to acquire original image information;

[0073] The preprocessing feature matrix acquisition module is used to preprocess the original image information to obtain a preprocessing feature matrix;

[0074] The network acquisition module is used to acquire the two-level cascade calcification inward migration recognition network, the intimal sheet recognition network, and the pseudo-CTA image generation network;

[0075] The calcification inward shift recognition image acquisition module is used to input the preprocessing feature matrix into the two-stage cascade calcification inward shift recognition network, thereby acquiring a calcification inward shift recognition image;

[0076] The endothelial sheet recognition image acquisition module is used to input the preprocessing feature matrix into the endothelial sheet recognition network, so as to acquire an endothelial sheet recognition image;

[0077] The pseudo CTA image acquisition module is used to input the preprocessed feature matrix into the pseudo CTA image generation network, so as to acquire a pseudo CTA image.

[0078] Figure 2 It is a structural block diagram of an electronic device provided by one or more embodiments of the present invention.

[0079] like Figure 2 As shown, the present application also discloses an electronic device, including: a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the aortic dissection image acquisition method based on non-enhanced CT images.

[0080] The present application also provides a computer-readable storage medium storing a computer program executable by an electronic device. When the computer program runs on the electronic device, the steps of the aortic dissection image acquisition method based on non-enhanced CT images can be implemented.

[0081] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0082] The electronic device includes a hardware layer, an operating system layer running on the hardware layer, and an application layer running on the operating system. The hardware layer includes hardware such as a central processing unit (CPU), a memory management unit (MMU), and a memory. The operating system can be any one or more computer operating systems that implement electronic device control through a process, such as a Linux operating system, a Unix operating system, an Android operating system, an iOS operating system, or a Windows operating system. In addition, in an embodiment of the present invention, the electronic device can be a handheld device such as a smart phone or a tablet computer, or can be an electronic device such as a desktop computer or a portable computer, which is not particularly limited in the embodiment of the present invention.

[0083] The execution subject of the electronic device control in the embodiment of the present invention may be an electronic device, or a functional module in the electronic device that can call and execute a program. The electronic device may obtain the firmware corresponding to the storage medium. The firmware corresponding to the storage medium is provided by the supplier. The firmware corresponding to different storage media may be the same or different, which is not limited here. After the electronic device obtains the firmware corresponding to the storage medium, the firmware corresponding to the storage medium may be written into the storage medium, specifically, the firmware corresponding to the storage medium may be burned into the storage medium. The process of burning the firmware into the storage medium may be implemented using existing technology, which will not be described in detail in the embodiment of the present invention.

[0084] The electronic device may also obtain a reset command corresponding to the storage medium. The reset command corresponding to the storage medium is provided by the supplier. The reset commands corresponding to different storage media may be the same or different, and are not limited here.

[0085] At this time, the storage medium of the electronic device is a storage medium in which the corresponding firmware is written, and the electronic device can respond to the reset command corresponding to the storage medium in the storage medium in which the corresponding firmware is written, so that the electronic device resets the storage medium in which the corresponding firmware is written according to the reset command corresponding to the storage medium. The process of resetting the storage medium according to the reset command can be implemented by the existing technology and will not be described in detail in the embodiments of the present invention.

[0086] For the convenience of description, the above devices are described in terms of functions and are divided into various units and modules. Of course, when implementing the present application, the functions of each unit and module can be implemented in the same or multiple software and / or hardware.

[0087] Those skilled in the art will appreciate that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined.

[0088] For the method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0089] It can be known from the description of the above implementation modes that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly contributed to the prior art in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute the methods described in the various implementation modes of the present application or certain parts of the implementation modes.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for acquiring an aortic dissection image based on a non-enhanced CT image, characterized in that: The aortic dissection image acquisition method based on non-enhanced CT images comprises: Get original image information; Preprocessing the original image information to obtain a preprocessing feature matrix; Obtain a two-level cascade calcification inward migration recognition network, an intimal sheet recognition network, and a pseudo-CTA image generation network; Inputting the preprocessed feature matrix into the two-stage cascade calcification inward shift recognition network, thereby obtaining a calcification inward shift recognition image; Inputting the preprocessed feature matrix into the endothelial sheet recognition network to obtain an endothelial sheet recognition image; The preprocessed feature matrix is ​​input into the pseudo CTA image generation network to obtain a pseudo CTA image.

2. The method for acquiring an aortic dissection image based on a non-enhanced CT image according to claim 1, characterized in that: The preprocessing of the original data to obtain a preprocessing matrix includes: The original data were windowed and the voxel spacing was normalized.

3. The method for acquiring an aortic dissection image based on a non-enhanced CT image according to claim 2, characterized in that: The two-stage cascade neural network comprises: The first-level NCCT aorta segmentation network adopts the nnUnet architecture; The second-level calcification inward migration segmentation network, the output of the first-level NCCT aorta segmentation network is used as the input of the second-level calcification inward migration segmentation network; wherein, The downsampling step size of the z-direction convolution kernel of the network backbone layer of the two-stage cascade neural network is changed to 1.

4. The method for acquiring an aortic dissection image based on a non-enhanced CT image according to claim 3, characterized in that: The inner membrane sheet recognition network includes a 3D segmentation network and a 2D segmentation network.

5. The method for acquiring an aortic dissection image based on a non-enhanced CT image according to claim 4, characterized in that: The pseudo CTA image generation network includes a generator, a discriminator and a registration network. The original image information is converted into a pseudo CTA image through the generator, and the original image information and the pseudo CTA image are respectively sent to the discriminator to obtain the results.

6. An aortic dissection image acquisition device based on non-enhanced CT images, characterized in that: The aortic dissection image acquisition device based on non-enhanced CT images comprises: An original image information acquisition module, wherein the original image information acquisition module is used to acquire original image information; A preprocessing feature matrix acquisition module, wherein the preprocessing feature matrix acquisition module is used to preprocess the original image information to obtain a preprocessing feature matrix; A network acquisition module, the network acquisition module is used to acquire a two-stage cascade calcification inward migration recognition network, an intimal sheet recognition network, and a pseudo CTA image generation network; A calcification inward shift recognition image acquisition module, the calcification inward shift recognition image acquisition module is used to input the preprocessing feature matrix into the two-stage cascade calcification inward shift recognition network, so as to acquire a calcification inward shift recognition image; An endothelial patch recognition image acquisition module, wherein the endothelial patch recognition image acquisition module is used to input the preprocessed feature matrix into the endothelial patch recognition network, thereby acquiring an endothelial patch recognition image; A pseudo CTA image acquisition module is used to input the preprocessed feature matrix into the pseudo CTA image generation network to acquire a pseudo CTA image.

Images