A method and apparatus for patch artifact correction and component analysis based on artificial intelligence

Abstract

This invention discloses an artificial intelligence-based method and apparatus for plaque artifact correction and component analysis, relating to the field of medical imaging technology. One specific embodiment of the method includes: acquiring an artifact image of a target region; repairing the artifact image using a pre-trained artifact correction model to obtain a corrected image corresponding to the artifact image; and detecting the corrected image and the artifact image based on a pre-trained plaque component discrimination model to obtain the component discrimination result of the plaque in the target region. This embodiment can improve the accuracy of plaque component detection.

Description

Technical Field

[0001] This invention relates to the field of medical imaging technology, and in particular to a method and apparatus for patch artifact correction and component analysis based on artificial intelligence. Background Technology

[0002] Analyzing the cost of vascular plaques is crucial for determining whether a patient will subsequently develop an embolism. Current techniques typically use raw images with artifacts to directly identify plaque components; however, due to the adverse effects of artifacts, the accuracy of this method is low. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a method and apparatus for patch artifact correction and component analysis based on artificial intelligence. By correcting the artifact image and judging the patch component based on the corrected image and the artifact image together, the accuracy of patch component detection is improved.

[0004] To achieve the above objectives, according to one aspect of the present invention, an artificial intelligence-based method for patch artifact correction and component analysis is provided.

[0005] The artificial intelligence-based patch artifact correction and component analysis method of this invention includes: acquiring an artifact image of a target region; repairing the artifact image using a pre-trained artifact correction model to obtain a corrected image corresponding to the artifact image; and detecting the corrected image and the artifact image according to a pre-trained patch component discrimination model to obtain the component discrimination result of the patches in the target region.

[0006] Optionally, the artifact correction model is a generative adversarial network, trained through the following steps: acquiring a normal image of the target region without artifacts and lesions; inputting the normal image into a pre-trained artifact simulation model to obtain a first image with artifacts; inputting the normal image into a pre-trained lesion simulation model to obtain a second image with lesions; inputting the first image into the lesion simulation model to obtain a third image with artifacts and lesions; training the artifact correction model using the third image as a training sample and the second image as the label of the training sample.

[0007] Optionally, both the artifact simulation model and the lesion simulation model are generative adversarial networks; wherein, the artifact simulation model is trained adversarially using artifact samples as training samples and artifact removal samples corresponding to the artifact samples as labels; and the lesion simulation model is trained using lesion samples as training samples and lesion-free samples corresponding to the lesion samples as labels.

[0008] Optionally, the artifact correction model is a generative adversarial network, trained through the following steps: acquiring a systolic image of the target region and a corresponding diastolic image; training the artifact correction model using the systolic image as a training sample and the diastolic image as the label of the training sample.

[0009] Optionally, the step of detecting the modified image and the artifact image based on a pre-trained patch composition discrimination model to obtain the composition discrimination result of the patches in the target region includes: inputting the modified image and the artifact image into the patch composition discrimination model to obtain the composition discrimination result of the patches in the target region.

[0010] Optionally, the step of detecting the modified image and the artifact image based on a pre-trained plaque component discrimination model to obtain the component discrimination result of the plaque in the target region includes: inputting the modified image, the artifact image, and the lesion image detected in advance from the artifact image into the plaque component discrimination model to obtain the component discrimination result of the plaque in the target region.

[0011] To achieve the above objectives, according to another aspect of the present invention, an artificial intelligence-based patch artifact correction and component analysis device is provided.

[0012] The artificial intelligence-based patch artifact correction and composition analysis device of this invention may include: an artifact acquisition unit for acquiring artifact images of a target region; an artifact correction unit for repairing the artifact images using a pre-trained artifact correction model to obtain a corrected image corresponding to the artifact images; and a detection unit for detecting the corrected image and the artifact image according to a pre-trained patch composition discrimination model to obtain the composition discrimination result of the patches in the target region.

[0013] Optionally, the detection unit may be further configured to: input the corrected image and the artifact image into the plaque composition discrimination model to obtain the composition discrimination result of the plaque in the target region; or, input the corrected image, the artifact image, and the lesion image pre-detected from the artifact image into the plaque composition discrimination model to obtain the composition discrimination result of the plaque in the target region.

[0014] To achieve the above objectives, according to another aspect of the present invention, an electronic device is provided.

[0015] An electronic device according to the present invention includes: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the artificial intelligence-based patch artifact correction and component analysis method provided by the present invention.

[0016] To achieve the above objectives, according to another aspect of the present invention, a computer-readable storage medium is provided.

[0017] The present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the artificial intelligence-based patch artifact correction and component analysis method provided by the present invention.

[0018] According to the technical solution of the present invention, the embodiments described above have the following advantages or beneficial effects:

[0019] After acquiring the artifact image of the target region, a pre-trained artifact correction model is first used to repair the artifact image, resulting in a corrected image. Then, a pre-trained plaque component discrimination model is used to detect both the corrected and artifact images, thereby obtaining the component discrimination result of the plaques in the target region. Through these steps, the accuracy of plaque component detection can be improved by using the corrected image after artifact removal. Furthermore, the corrected image, artifact image, and lesion images pre-detected from the artifact image can be input together into the plaque component discrimination model, which helps to obtain a more accurate plaque component discrimination result. Furthermore, this invention also provides an effective method for training a plaque component discrimination model. Specifically, firstly, a normal image of the target region without artifacts or lesions is acquired; then, the normal image is input into a pre-trained artifact simulation model to obtain a first image with artifacts; the normal image is input into a pre-trained lesion simulation model to obtain a second image with lesions; the first image is input into the lesion simulation model to obtain a third image with both artifacts and lesions; finally, an artifact correction model is trained using the third image as a training sample and the second image as a training sample. This training method can generate the necessary images for training through a machine learning model, thus requiring fewer original images while achieving better model training results.

[0020] The further effects of the aforementioned unconventional alternative methods will be explained below in conjunction with specific implementation methods. Attached Figure Description

[0021] The accompanying drawings are provided to better understand the invention and are not intended to unduly limit the scope of the invention. Wherein:

[0022] Figure 1 This is a schematic diagram of the main steps of the artificial intelligence-based patch artifact correction and component analysis method in this embodiment of the invention;

[0023] Figure 2 This is a schematic diagram illustrating the training and usage steps of the artifact correction model in an embodiment of the present invention;

[0024] Figure 3This is a schematic diagram of the components of the artificial intelligence-based patch artifact correction and component analysis device in an embodiment of the present invention;

[0025] Figure 4 This is an exemplary system architecture diagram that can be applied thereto according to embodiments of the present invention;

[0026] Figure 5 This is a schematic diagram of the electronic device structure used to implement the artificial intelligence-based patch artifact correction and component analysis method in the embodiments of the present invention. Detailed Implementation

[0027] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of the present invention, including various details to aid understanding. These details should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0028] It should be noted that, unless otherwise specified, the embodiments of the present invention and the technical features thereof can be combined with each other.

[0029] Figure 1 This is a schematic diagram of the main steps of the patch artifact correction and component analysis method based on artificial intelligence according to an embodiment of the present invention.

[0030] like Figure 1 As shown, the artificial intelligence-based patch artifact correction and component analysis method of this invention can be specifically executed according to the following steps:

[0031] Step S101: Obtain the artifact image of the target region.

[0032] In this embodiment of the invention, the target area can be any region of the human body to be determined, such as the coronary artery region or the cerebral artery region. The following description uses the coronary artery region as the target area. The artifact images mentioned above refer to the original images acquired by medical testing equipment such as CT (Computed Tomography). It can be understood that the original images generally contain artifacts, which are various forms of images that appear on the image but do not actually exist.

[0033] Step S102: Repair the artifact image using a pre-trained artifact correction model to obtain the corrected image corresponding to the artifact image.

[0034] In this step, the artifact correction model is a machine learning model used to remove artifacts from the artifact image and thus repair the artifact image. Preferably, the artifact correction model can be a Generative Adversarial Network (GAN).

[0035] Figure 2 This is a schematic diagram illustrating the training and usage steps of the artifact correction model in this embodiment of the invention, as shown below. Figure 2 As shown, the training steps of the artifact correction model are as follows: First, obtain a normal image of the target region without artifacts or lesions; then, input the normal image into a pre-trained artifact simulation model to obtain a first image with artifacts; input the normal image into a pre-trained lesion simulation model to obtain a second image with lesions; input the first image into the lesion simulation model to obtain a third image with both artifacts and lesions; finally, train the artifact correction model using the third image as the training sample and the second image as the label of the training sample. This training method can generate the necessary images for training through a machine learning model, thus requiring fewer original images, while achieving good model training results.

[0036] As a preferred embodiment, both the artifact simulation model and the lesion simulation model are generative adversarial networks (GANs). Specifically, the artifact simulation model can be trained adversarially using artifact samples as training samples and artifact removal samples corresponding to the artifact samples as labels; the lesion simulation model can be trained using lesion samples as training samples and lesion-free samples corresponding to the lesion samples as labels.

[0037] In an optional technical solution, the artifact correction model can also be trained through the following steps: acquiring a systolic image of the target region and a corresponding diastolic image; training the artifact correction model using the systolic image as a training sample and the diastolic image as the label of the training sample. It is understood that for corresponding systolic and diastolic images, the diastolic image often removes artifacts at the same location in the systolic image; therefore, both can be used as training data for the artifact correction model.

[0038] After the artifact correction model is trained, the artifact image from step S101 can be input into the artifact correction model to obtain the corrected image corresponding to the artifact image.

[0039] Step S103: Detect the corrected image and the artifact image based on the pre-trained patch composition discrimination model to obtain the composition discrimination result of the patches in the target region.

[0040] In this step, the plaque component discrimination model is a machine learning model used to detect the components of vascular plaques. It can generally include a convolutional neural network (CNN) for extracting image features and a classification network for detecting plaque components based on image features.

[0041] In practical applications, plaque composition can be obtained in two ways. The first method involves inputting the corrected image and the artifact image into the plaque composition discrimination model to obtain the composition discrimination result of the plaque in the target region. For example, the plaque composition discrimination result can be: whether it contains calcification, lipids, fibrous components, and the proportion of each component. The second method involves inputting the corrected image, the artifact image, and a lesion image (i.e., an image containing plaques or other lesions) pre-detected from the artifact image into the plaque composition discrimination model to obtain the composition discrimination result of the plaque in the target region. The lesion image can be obtained through the following steps: first, using a pre-determined lesion detection algorithm to determine the lesion area from the artifact image; then, extracting the image of the lesion area from the artifact image to obtain the lesion image. It can be understood that because the lesion image focusing on the plaque is added to the plaque composition discrimination model for detection, the second method has higher accuracy in plaque composition detection.

[0042] According to the technical solution of this invention, after acquiring the artifact image of the target region, the artifact image is first repaired using a pre-trained artifact correction model to obtain a corrected image corresponding to the artifact image; then, the corrected image and the artifact image are detected based on a pre-trained plaque component discrimination model to obtain the component discrimination result of the plaque in the target region. Through the above steps, the accuracy of plaque component detection can be improved by using the corrected image after removing artifacts. Furthermore, the corrected image, the artifact image, and the lesion image detected in advance from the artifact image can be input together into the plaque component discrimination model, which helps to obtain a more accurate plaque component discrimination result. Furthermore, this invention also provides an effective method for training a plaque component discrimination model. Specifically, firstly, a normal image of the target region without artifacts or lesions is acquired; then, the normal image is input into a pre-trained artifact simulation model to obtain a first image with artifacts; the normal image is input into a pre-trained lesion simulation model to obtain a second image with lesions; the first image is input into the lesion simulation model to obtain a third image with both artifacts and lesions; finally, an artifact correction model is trained using the third image as a training sample and the second image as a training sample. This training method can generate the necessary images for training through a machine learning model, thus requiring fewer original images while achieving better model training results.

[0043] It should be noted that, for the sake of ease of description, the foregoing method embodiments are described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, and some steps may actually be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential for implementing the present invention.

[0044] To facilitate better implementation of the above-described solutions of the embodiments of the present invention, related apparatus for implementing the above-described solutions is also provided below.

[0045] Please see Figure 3 As shown, the artificial intelligence-based patch artifact correction and component analysis device 300 provided in this embodiment of the invention may include: an artifact acquisition unit 301, an artifact correction unit 302, and a detection unit 303.

[0046] The artifact acquisition unit 301 is used to acquire artifact images of the target region; the artifact correction unit 302 is used to repair the artifact image using a pre-trained artifact correction model to obtain a corrected image corresponding to the artifact image; and the detection unit 303 is used to detect the corrected image and the artifact image according to a pre-trained patch composition discrimination model to obtain the composition discrimination result of the patches in the target region.

[0047] In this embodiment of the invention, the detection unit 303 may be further configured to: input the corrected image and the artifact image into the plaque component discrimination model to obtain the component discrimination result of the plaque in the target region; or, input the corrected image, the artifact image, and the lesion image detected in advance from the artifact image into the plaque component discrimination model to obtain the component discrimination result of the plaque in the target region.

[0048] In a specific application, the artifact correction model is a generative adversarial network; the device 300 may further include a training unit, which is used to: acquire a normal image of the target region without artifacts and without lesions; input the normal image into a pre-trained artifact simulation model to obtain a first image with artifacts; input the normal image into a pre-trained lesion simulation model to obtain a second image with lesions; input the first image into the lesion simulation model to obtain a third image with artifacts and lesions; and train the artifact correction model using the third image as a training sample and the second image as the label of the training sample.

[0049] As a preferred embodiment, both the artifact simulation model and the lesion simulation model are generative adversarial networks; wherein, the artifact simulation model is trained adversarially using artifact samples as training samples and artifact removal samples corresponding to the artifact samples as labels; and the lesion simulation model is trained using lesion samples as training samples and lesion-free samples corresponding to the lesion samples as labels.

[0050] Furthermore, in this embodiment of the invention, the training unit can be further used to: acquire a systolic image of the target region and a diastolic image corresponding to the systolic image; and train the artifact correction model using the systolic image as a training sample and the diastolic image as the label of the training sample.

[0051] According to the technical solution of this invention, after acquiring the artifact image of the target region, the artifact image is first repaired using a pre-trained artifact correction model to obtain a corrected image corresponding to the artifact image; then, the corrected image and the artifact image are detected based on a pre-trained plaque component discrimination model to obtain the component discrimination result of the plaque in the target region. Through the above steps, the accuracy of plaque component detection can be improved by using the corrected image after removing artifacts. Furthermore, the corrected image, the artifact image, and the lesion image detected in advance from the artifact image can be input together into the plaque component discrimination model, which helps to obtain a more accurate plaque component discrimination result. Furthermore, this invention also provides an effective method for training a plaque component discrimination model. Specifically, firstly, a normal image of the target region without artifacts or lesions is acquired; then, the normal image is input into a pre-trained artifact simulation model to obtain a first image with artifacts; the normal image is input into a pre-trained lesion simulation model to obtain a second image with lesions; the first image is input into the lesion simulation model to obtain a third image with both artifacts and lesions; finally, an artifact correction model is trained using the third image as a training sample and the second image as a training sample. This training method can generate the necessary images for training through a machine learning model, thus requiring fewer original images while achieving better model training results.

[0052] Figure 4 An exemplary system architecture 400 is shown that can be applied to the AI-based patch artifact correction and component analysis method or the AI-based patch artifact correction and component analysis apparatus of the present invention.

[0053] like Figure 4As shown, system architecture 400 may include terminal devices 401, 402, and 403, network 404, and server 405 (this architecture is merely an example; the components included in a specific architecture may be adjusted according to the specific application). Network 404 serves as the medium for providing a communication link between terminal devices 401, 402, and 403 and server 405. Network 404 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0054] Users can use terminal devices 401, 402, and 403 to interact with server 405 via network 404 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 401, 402, and 403, such as plaque composition identification applications (for example only).

[0055] Terminal devices 401, 402, and 403 can be various electronic devices with displays that support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0056] Server 405 can be a server that provides various services, such as a backend server that supports the patch composition discrimination application operated by the user using terminal devices 401, 402, and 403 (for example only). The backend server can process received patch detection requests and feed back the processing results (such as the specific patch components detected - for example only) to terminal devices 401, 402, and 403.

[0057] It should be noted that the artificial intelligence-based patch artifact correction and component analysis method provided in the embodiments of the present invention is generally executed by server 405, and correspondingly, the artificial intelligence-based patch artifact correction and component analysis device is generally set in server 405.

[0058] It should be understood that Figure 4 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0059] The present invention also provides an electronic device. The electronic device of this invention includes: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement the artificial intelligence-based patch artifact correction and component analysis method provided by the present invention.

[0060] The following is for reference. Figure 5 It shows a schematic diagram of the structure of a computer system 500 suitable for implementing an electronic device according to embodiments of the present invention. Figure 5The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0061] like Figure 5 As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 502 or programs loaded from storage section 508 into random access memory (RAM) 503. The RAM 503 also stores various programs and data required for the operation of the computer system 500. The CPU 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0062] The following components are connected to I / O interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to I / O interface 505 as needed. A removable medium 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 510 as needed so that computer programs read from it can be installed into storage section 508 as needed.

[0063] In particular, according to the embodiments disclosed in this invention, the processes described in the above main step diagrams can be implemented as computer software programs. For example, embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the main step diagrams. In the above embodiments, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is executed by central processing unit 501, it performs the functions defined in the system of this invention.

[0064] It should be noted that the computer-readable medium shown in this invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0065] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0066] The units described in the embodiments of the present invention can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor can be described as including an artifact acquisition unit, an artifact correction unit, and a detection unit. The names of these units do not necessarily limit the specific unit; for example, the artifact acquisition unit can also be described as "a unit that provides artifact images to the artifact correction unit."

[0067] In another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiments; or it may exist independently and not assembled into the device. The computer-readable medium carries one or more programs, which, when executed by the device, cause the device to perform the following steps: acquiring an artifact image of a target region; repairing the artifact image using a pre-trained artifact correction model to obtain a corrected image corresponding to the artifact image; and detecting the corrected image and the artifact image according to a pre-trained patch composition discrimination model to obtain a patch composition discrimination result in the target region.

[0068] According to the technical solution of this invention, after acquiring the artifact image of the target region, the artifact image is first repaired using a pre-trained artifact correction model to obtain a corrected image corresponding to the artifact image; then, the corrected image and the artifact image are detected based on a pre-trained plaque component discrimination model to obtain the component discrimination result of the plaque in the target region. Through the above steps, the accuracy of plaque component detection can be improved by using the corrected image after removing artifacts. Furthermore, the corrected image, the artifact image, and the lesion image detected in advance from the artifact image can be input together into the plaque component discrimination model, which helps to obtain a more accurate plaque component discrimination result. Furthermore, this invention also provides an effective method for training a plaque component discrimination model. Specifically, firstly, a normal image of the target region without artifacts or lesions is acquired; then, the normal image is input into a pre-trained artifact simulation model to obtain a first image with artifacts; the normal image is input into a pre-trained lesion simulation model to obtain a second image with lesions; the first image is input into the lesion simulation model to obtain a third image with both artifacts and lesions; finally, an artifact correction model is trained using the third image as a training sample and the second image as a training sample. This training method can generate the necessary images for training through a machine learning model, thus requiring fewer original images while achieving better model training results.

[0069] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for patch artifact correction and component analysis based on artificial intelligence, characterized in that, include: Obtain artifact images of the target region; The artifact image is repaired using a pre-trained artifact correction model to obtain a corrected image corresponding to the artifact image; wherein, the corrected image is obtained after removing artifacts from the artifact image; The modified image and the artifact image are detected based on a pre-trained plaque component discrimination model to obtain the component discrimination result of the plaque in the target region. This includes: inputting the modified image and the artifact image into the plaque component discrimination model, or inputting the modified image, the artifact image, and a lesion image pre-detected from the artifact image into the plaque component discrimination model; and obtaining the component discrimination result of the plaque in the target region. The artifact correction model is a generative adversarial network, trained through the following steps: acquiring a normal image of the target region without artifacts or lesions; inputting the normal image into a pre-trained artifact simulation model to obtain a first image with artifacts; inputting the normal image into a pre-trained lesion simulation model to obtain a second image with lesions; inputting the first image into the lesion simulation model to obtain a third image with artifacts and lesions; training the artifact correction model using the third image as a training sample and the second image as the label of the training sample.

2. The method according to claim 1, characterized in that, Both the artifact simulation model and the lesion simulation model are generative adversarial networks; among them... The artifact simulation model is trained using artifact samples as training samples and artifact removal samples corresponding to the artifact samples as labels for adversarial training. The lesion simulation model is trained using lesion samples as training samples and corresponding lesion-free samples as labels.

3. The method according to claim 1, characterized in that, The artifact correction model is a generative adversarial network, trained through the following steps: Acquire the systolic image of the target region and the corresponding diastolic image; The artifact correction model is trained using the systolic images as training samples and the diastolic images as labels for the training samples.

4. A patch artifact correction and component analysis device based on artificial intelligence, characterized in that, include: The artifact acquisition unit is used to acquire artifact images of the target area. The artifact correction unit is used to repair the artifact image using a pre-trained artifact correction model to obtain a corrected image corresponding to the artifact image; wherein the corrected image is obtained after removing artifacts from the artifact image; The detection unit is used to detect the corrected image and the artifact image based on a pre-trained patch composition discrimination model, and obtain the composition discrimination result of the patches in the target region; The detection unit is further configured to: input the corrected image and the artifact image into the plaque component discrimination model, or input the corrected image, the artifact image, and the lesion image pre-detected from the artifact image into the plaque component discrimination model; and obtain the component discrimination result of the plaque in the target region; The device further includes a training unit, configured to: acquire a normal image of the target region without artifacts and lesions; input the normal image into a pre-trained artifact simulation model to obtain a first image with artifacts; input the normal image into a pre-trained lesion simulation model to obtain a second image with lesions; input the first image into the lesion simulation model to obtain a third image with artifacts and lesions; and train the artifact correction model using the third image as a training sample and the second image as the label of the training sample.

5. An electronic device, characterized in that, include: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-3.

Images