Image artifact removal methods, systems, devices, and media

Abstract

The application discloses an image artifact removal method, system, device and medium, and relates to the technical field of image processing. The image artifact removal method comprises the following steps: acquiring an initial mask image, an initial subtraction image, an initial contrast image, and an initial mask reconstruction image obtained by reconstructing the initial mask image; performing segmentation processing and projection processing on the initial mask reconstruction image to obtain a mask projection image marking a target region; performing screening processing on the initial subtraction image based on the mask projection image to obtain an intermediate subtraction image; performing image reconstruction processing based on the intermediate subtraction image to obtain a target subtraction reconstruction image; performing image reconstruction processing based on the initial mask reconstruction image and the mask projection image to determine a target mask reconstruction image; and performing image fusion processing based on the target mask reconstruction image and the target subtraction reconstruction image to obtain a target image after artifact removal.

Description

Technical Field

[0001] This application relates to the technical fields of image processing, and in particular to a method, system, device and medium for removing image artifacts. Background Technology

[0002] When using radiological diagnostic instruments such as CT (Computed Tomography) to scan patient sites containing metallic implants with X-rays, the high absorption of X-rays by high-density materials like metal results in metallic artifacts in the reconstructed image, with alternating bright and dark areas. Furthermore, the sharp decrease in photon count as X-rays pass through metal leads to a very weak signal received by the detector, producing strong noise and streaks. These artifacts severely affect image quality, reducing the visual appeal of the generated image and impacting diagnostic accuracy. In 3DDSA (Digital Subtraction Angiography), when a metallic substance is implanted in a patient, the subtraction of the first mask image and the second contrast image to obtain the vascular sub-image can lead to residual metal due to patient movement, gantry instability, or misregistration. This can cause inaccurate fusion, giving the doctor the false impression of recurrence, such as a hemangioma, thus affecting the diagnosis.

[0003] Related technologies are based on traditional classical methods in the projection domain. A representative algorithm is LI-MAR (Linear Interpolation Metal Artifact Reduction), which reconstructs an image containing initial metal artifacts from the original projection data. Metal segmentation is then performed to generate a metal trajectory map. Linear interpolation is then used to repair the contaminated data, resulting in the final reconstructed and corrected image. The NLI-MAR algorithm (Normalized Linear Interpolation Metal Artifact Reduction) first creates a prior image that maps the attenuation values ​​of all tissues (including bone) in the original image to a more uniform range closer to soft tissue. Interpolation and repair are then performed on this normalized image, avoiding the crude processing of peak data. Finally, inverse normalization is used to restore the true tissue contrast. However, the above algorithms are highly dependent on segmentation accuracy. If the initial metal segmentation is inaccurate, the contaminated area will be incorrectly marked, resulting in artifact residue or normal tissue being incorrectly corrected. Furthermore, the computational cost of finding an image that best matches the forward projection and the measured effective projection data (non-metallic area) based on the iterative reconstruction method is extremely high. Summary of the Invention

[0004] The embodiments of this application aim to at least partially solve one of the technical problems in the related art. Therefore, the purpose of the embodiments of this application is to provide an image artifact removal method, system, apparatus, and medium that improves the quality of the output image.

[0005] This application provides an image artifact removal method, comprising: acquiring an initial mask image, an initial subtraction image, an initial contrast image, and an initial mask reconstruction image obtained by reconstructing the initial mask image, wherein the initial subtraction image is obtained by subtraction processing based on the initial mask image and the initial contrast image; performing segmentation and projection processing on the initial mask reconstruction image to obtain a mask projection image marking the target region; performing filtering processing on the initial subtraction image based on the mask projection image to obtain an intermediate subtraction image; performing image reconstruction processing on the intermediate subtraction image to obtain a target subtraction reconstruction image; performing image reconstruction processing on the initial mask reconstruction image and the mask projection image to determine the target mask reconstruction image; and performing image fusion processing on the target mask reconstruction image and the target subtraction reconstruction image to obtain a target image after artifact removal.

[0006] For example, segmenting and projecting an initial mask reconstruction image to obtain a mask projection image marking the target region includes: segmenting the initial mask reconstruction image based on a first threshold to mark the target region of the initial mask reconstruction image; performing erosion and dilation processing on the initial mask reconstruction image marking the target region to update the target region in the initial mask reconstruction image, obtaining an erosion-dilated initial mask reconstruction image; and projecting the erosion-dilated initial mask reconstruction image based on a ray-driven model to obtain a mask projection image marking the target region, wherein the ray-driven model is implemented based on dual rays, and the rays are emitted when the device scans and acquires the image.

[0007] For example, filtering an initial subtraction image based on a mask projection image to obtain an intermediate subtraction image includes: determining the afterimage region of the initial subtraction image based on the target region of the mask projection image; setting a second threshold, and for pixels in the afterimage region, if the pixel grayscale value is greater than the second threshold and the number of pixels with grayscale values ​​greater than the second threshold in the connected component to which the pixel belongs is greater than the set threshold, retaining its pixel grayscale value; otherwise, setting the pixel grayscale value to a first preset value to obtain the intermediate subtraction image.

[0008] For example, determining a target mask reconstruction image by performing image reconstruction processing based on an initial mask reconstruction image and a mask projection image includes: interpolating the target region in the mask projection image to obtain a first projection image; performing image reconstruction processing on the first projection image to obtain a first mask reconstruction image; binarizing the target region in the mask projection image to obtain a second projection image; performing image reconstruction processing on the second projection image to obtain a second mask reconstruction image; and determining the target mask reconstruction image based on the initial mask reconstruction image, the first mask reconstruction image, and the second mask reconstruction image.

[0009] For example, binarizing the target region in the mask projection image to obtain a second projection image includes: judging the gray value of each pixel corresponding to the target region in the mask projection image based on a third threshold to obtain the second projection image, wherein when the gray value is greater than the third threshold, the gray value is set to a second preset value, and when the gray value is less than or equal to the third threshold, the gray value is set to a third preset value. For example, the second mask reconstruction image includes multiple slice sequences; determining the target mask reconstruction image based on the initial mask reconstruction image, the first mask reconstruction image, and the second mask reconstruction image includes: obtaining a reference scanning angle for device scanning; determining the number of pixels with grayscale values ​​of a second preset value in each slice sequence of the second mask reconstruction image, and obtaining the scanning angle of each pixel based on the number; and determining the target mask reconstruction image based on the scanning angle, the reference scanning angle, the angle threshold, the initial mask reconstruction image, and the first mask reconstruction image.

[0010] For example, determining the target mask reconstruction image based on the scanning angle, reference scanning angle, angle threshold, initial mask reconstruction image, and first mask reconstruction image includes: judging each pixel of the first mask reconstruction image based on the scanning angle, reference scanning angle, and angle threshold; when the scanning angle is greater than the angle threshold and less than the reference scanning angle, interpolating the first mask reconstruction image based on the initial mask reconstruction image to obtain the target mask reconstruction image corresponding to the pixel; when the scanning angle is equal to the reference scanning angle, using the gray value of the initial mask reconstruction image corresponding to the pixel as the gray value of the pixel in the first mask reconstruction image, and obtaining the target mask reconstruction image corresponding to the pixel based on the first mask reconstruction image.

[0011] For example, interpolating the first mask reconstruction image based on the initial mask reconstruction image to obtain the target mask reconstruction image includes: when the scanning angle of a pixel is greater than an angle threshold and less than a reference scanning angle, determining a first difference between the reference scanning angle and the angle threshold, a second difference between the reference scanning angle and the scanning angle, and a third difference between the scanning angle and the angle threshold; determining a first ratio between the second difference and the first difference, and multiplying the gray value of the first mask reconstruction image corresponding to the pixel with the first ratio to obtain a first product result; determining a second ratio between the third difference and the first difference, and multiplying the gray value of the initial mask reconstruction image corresponding to the pixel with the second ratio to obtain a second product result; performing an addition operation based on the first product result and the second product result to obtain the gray value of the pixel in the first mask reconstruction image, and obtaining the target mask reconstruction image corresponding to the pixel based on the first mask reconstruction image.

[0012] Another embodiment of this application provides an image artifact removal system, comprising: an acquisition module for acquiring an initial mask image, an initial subtraction image, an initial contrast image, and an initial mask reconstruction image obtained by reconstructing the initial mask image, wherein the initial subtraction image is obtained by subtraction processing based on the initial mask image and the initial contrast image; a first processing module for performing segmentation and projection processing on the initial mask reconstruction image to obtain a mask projection image of a marked target region; a second processing module for performing filtering processing on the initial subtraction image based on the mask projection image to obtain an intermediate subtraction image; a first reconstruction module for performing image reconstruction processing based on the intermediate subtraction image to obtain a target subtraction reconstruction image; a second reconstruction module for performing image reconstruction processing based on the initial mask reconstruction image and the mask projection image to determine the target mask reconstruction image; and a fusion module for performing image fusion processing based on the target mask reconstruction image and the target subtraction reconstruction image to obtain a target image after artifact removal.

[0013] Another embodiment of this application provides an electronic device having a computer program stored thereon, which, when executed by a processor, implements the steps of the method of any of the above embodiments.

[0014] Another embodiment of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method of any of the above embodiments.

[0015] In the above embodiments, the image artifact removal method includes: acquiring an initial mask image, an initial subtraction image, an initial contrast image, and an initial mask reconstruction image obtained by reconstructing the initial mask image, wherein the initial subtraction image is obtained by subtraction processing based on the initial mask image and the initial contrast image; performing segmentation and projection processing on the initial mask reconstruction image to obtain a mask projection image marking the target region; performing filtering processing on the initial subtraction image based on the mask projection image to obtain an intermediate subtraction image; performing image reconstruction processing on the intermediate subtraction image to obtain a target subtraction reconstruction image; performing image reconstruction processing on the initial mask reconstruction image and the mask projection image to determine the target mask reconstruction image; and performing image fusion processing on the target mask reconstruction image and the target subtraction reconstruction image to obtain the target image after artifact removal. By sequentially acquiring the initial mask image, initial subtraction image, initial contrast image, and initial mask reconstruction image, and using the subtraction operation between the initial mask image and the initial contrast image to obtain the initial subtraction image, the initial mask reconstruction image is then segmented and projected to accurately generate a mask projection image that can mark the target region. A second subtraction operation is then performed on the initial subtraction image using the mask projection image to obtain an intermediate subtraction image. Subsequently, the first mask reconstruction image, the target subtraction reconstruction image, and the second mask reconstruction image are reconstructed by combining the mask projection image and the intermediate subtraction image. The optimal target mask reconstruction image is determined through joint screening of multiple frames of mask reconstruction images and fused with the target subtraction reconstruction image to reconstruct the final image. This method can effectively and accurately locate the distribution area of ​​artifacts such as metal, efficiently remove artifact interference components, significantly suppress stripe artifacts, scattering artifacts, and boundary distortion artifacts generated during imaging, significantly improve the imaging clarity and edge contour integrity of the target tissue structure, and effectively avoid the defects of traditional interpolation-based artifact removal algorithms that easily introduce secondary artifacts, loss of detail, and grayscale distortion. It enhances the overall signal-to-noise ratio and anatomical detail recognition of the image, ensuring the accuracy and reliability of subsequent clinical diagnosis and image analysis. Attached Figure Description

[0016] Figure 1 A basic flowchart of 3DDSA provided for the implementation of this application; Figure 2 A flowchart of an image artifact removal method provided for embodiments of this application; Figure 3 Flowchart of an image artifact removal method provided in another embodiment of this application; Figure 4 Comparison images of metal ghosting removal in sub-images provided for embodiments of this application; Figure 5 Comparison images of mask image before and after metal artifact removal provided for embodiments of this application; Figure 6 A block diagram of an image artifact removal system provided for another embodiment of this application; Figure 7 A block diagram of an electronic device provided for another embodiment of this application. Detailed Implementation

[0017] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0018] When using CT scanners or other radiological diagnostic instruments to perform X-ray scans on patient areas containing metallic implants, the reconstructed images often exhibit alternating bright and dark metallic artifacts due to the high absorption of X-rays by high-density materials like metal. Furthermore, the sharp decrease in photon count as X-rays pass through metal results in a very weak signal received by the detector, producing strong noise and streaks. These artifacts severely impact image quality, reducing the doctor's visual experience of the generated image and affecting diagnostic accuracy.

[0019] When 3DDSA ( Figure 1 In the basic flowchart of 3DDSA (Digital Subtraction Angiography), when a metal substance is implanted in a patient, the subtraction image of the blood vessel is obtained by subtracting the first mask image from the second contrast image. Due to patient movement, gantry instability, misregistration, etc., residual metal may remain, causing inaccurate fusion. This can give doctors the false impression of recurrence such as hemangioma, affecting the patient's diagnosis.

[0020] Related technologies are based on traditional classical methods in the projection domain. A representative algorithm is LI-MAR (Linear Interpolation Metal Artifact Reduction), which reconstructs an image containing initial metal artifacts from the original projection data. Metal segmentation is then performed to generate a metal trajectory map. Linear interpolation is then used to repair the contaminated data, resulting in the final reconstructed and corrected image. The NLI-MAR algorithm (Normalized Linear Interpolation Metal Artifact Reduction) first creates a prior image that maps the attenuation values ​​of all tissues (including bone) in the original image to a more uniform range closer to soft tissue. Interpolation and repair are then performed on this normalized image, avoiding the crude processing of peak data. Finally, inverse normalization is used to restore the true tissue contrast. However, the above algorithms are highly dependent on segmentation accuracy. If the initial metal segmentation is inaccurate, the contaminated area will be incorrectly marked, resulting in artifact residue or normal tissue being incorrectly corrected. Furthermore, the computational cost of finding an image that best matches the forward projection and the measured effective projection data (non-metallic area) based on the iterative reconstruction method is extremely high.

[0021] To address the issue of extreme reliance on segmentation performance, this application proposes an image artifact removal method. Since artifacts are caused by metal contamination of projection data, the method involves locating the metal region in the projection domain, interpolating the metal region to obtain an image without metal, and then reconstructing it to obtain a reconstructed image without metal artifacts. Additionally, it removes and reconstructs metal artifacts in the sub-images of 3DDSA, and performs image fusion to obtain a target image with metal artifacts removed. The image artifact removal method includes image input, metal artifact removal, and output of the image after artifact removal.

[0022] Figure 2 A flowchart illustrating the image artifact removal method provided in this application.

[0023] like Figure 2 As shown, the image artifact removal method 200 provided in this application embodiment includes, for example, steps S210-S260.

[0024] Step S210: Obtain the initial mask image, the initial subtraction image, the initial contrast image, and the initial mask reconstruction image obtained by reconstructing the initial mask image, wherein the initial subtraction image is obtained by subtraction processing based on the initial mask image and the initial contrast image. For example, based on the 3DDSA acquisition protocol (a set of fixed parameters and procedures that the device must strictly follow when performing a three-dimensional rotational DSA to ensure that the angle, number of frames, dose, and contrast agent injection sequence are consistent for each scan), a DSA device is used to scan to obtain the initial mask image (mask image, obtained by rotational scanning without contrast agent injection), the initial subtraction image (sub image, obtained by digital subtraction based on the mask image and contrast image), the initial contrast image (contrast image, obtained by scanning after contrast agent injection), and the initial mask reconstruction image (obtained by reconstructing the mask image, including multiple slice sequences).

[0025] Step S220: Perform segmentation and projection processing on the initial mask reconstruction image to obtain a mask projection image of the marked target region. For example, segmentation is performed based on a threshold segmentation method (first threshold), but it can also be performed using methods such as region growing and deep learning. Segmentation can mark the soft tissue and metal regions, and corrosion dilation is used to further optimize the boundaries between these regions, resulting in an optimized initial mask reconstruction image. This optimized initial mask reconstruction image is then projected to obtain a mask projection image of the marked target region (metal region). The projection process is implemented using a ray-driven model, specifically a dual-ray model where the rays are emitted during image acquisition by the 3DDSA device.

[0026] Step S230: The initial subtraction image is filtered based on the mask projection image to obtain the intermediate subtraction image. For example, based on the target area in the mask projection image, the corresponding afterimage area in the initial subtraction image can be determined; a second threshold (e.g., greater than the blood vessel value) is set, and for pixels in the afterimage area, the pixel gray value is retained only if the pixel gray value is greater than the second threshold and the number of pixels with gray values ​​greater than the second threshold in the connected region to which the pixel belongs is greater than the set threshold (e.g., 20). Otherwise, the pixel gray value is set to a first preset value (non-blood vessel value, e.g., set to 0, a negative value, or a constant much lower than the gray value of normal blood vessels, so that the pixel will not be regarded as a blood vessel signal in subsequent display or processing), thereby obtaining an intermediate subtraction image that removes isolated metal afterimages and retains the connected regions of blood vessels.

[0027] Step S240: Perform image reconstruction processing based on the intermediate subtraction image to obtain the target subtraction reconstructed image. For example, image reconstruction processing is performed on the intermediate subtraction image to obtain the target subtraction reconstructed image (a slice sequence of metal-free regions and metal artifacts).

[0028] Step S250: Perform image reconstruction processing based on the initial mask reconstruction image and the mask projection image to determine the target mask reconstruction image. For example, linear interpolation is performed on the metal region (target region) of the mask projection image to obtain the interpolated first projection image, and the target region in the mask projection image is binarized (for example, the gray value of each pixel corresponding to the target region in the mask projection image is judged, and when the gray value is greater than the third threshold, the gray value is set to the second preset value (e.g., 1), and when the gray value is less than or equal to the third threshold, the gray value is set to the third preset value (e.g., 0)) to obtain the second projection image. Image reconstruction processing is performed on the first projection image and the second projection image to obtain the first mask reconstruction image and the second mask reconstruction image. Obtain the reference scanning angle of the 3DDSA device (preset by the device), and determine the scanning angle of each pixel in the second mask reconstruction image (including multiple slice sequences) (for example, count the number of pixels with grayscale values ​​of 1 in each slice sequence, and use the sum of the numbers as the scanning angle of each pixel in the slice sequence). Compare the scanning angle of each pixel in the first mask reconstruction image with the reference scanning angle and the angle threshold. When the scanning angle is greater than the angle threshold and less than the reference scanning angle, interpolate the first mask reconstruction image (slice sequence without metal regions and metal artifacts) based on the initial mask reconstruction image (slice sequence with metal regions and metal artifacts) to obtain the target mask reconstruction image (slice sequence with metal regions and no metal artifacts) corresponding to the pixel. When the scanning angle is equal to the reference scanning angle, use the grayscale value of the initial mask reconstruction image corresponding to the pixel as the grayscale value of the pixel in the first mask reconstruction image, and obtain the target mask reconstruction image corresponding to the pixel based on the first mask reconstruction image. When the scanning angle is less than or equal to the angle threshold, directly use the grayscale value of the pixel in the first mask reconstruction image as the target mask reconstruction image of the pixel.

[0029] Step S260: Perform image fusion processing based on the target mask reconstruction image and the target subtraction reconstruction image to obtain the target image after artifact removal.

[0030] For example, the target mask reconstruction image (slice sequence) and the target subtraction reconstruction image (slice sequence) are fused to obtain a 3D target image after artifact removal.

[0031] In the above embodiments, an initial mask image, an initial subtraction image, an initial contrast image, and an initial mask reconstruction image are acquired sequentially. An initial subtraction image is obtained by subtracting the initial mask image from the initial contrast image. Then, the initial mask reconstruction image is segmented and projected to accurately generate a mask projection image of the target region. A second subtraction operation is performed on the initial subtraction image using the mask projection image to obtain an intermediate subtraction image. Subsequently, the first mask reconstruction image, the target subtraction reconstruction image, and the second mask reconstruction image are reconstructed by combining the mask projection image and the intermediate subtraction image. This process relies on multi-frame mask reconstruction. Joint image screening determines the optimal target mask reconstruction image and fuses it with the target subtraction reconstruction image to reconstruct the final image. This can effectively and accurately locate the distribution area of ​​artifacts such as metal, efficiently remove artifact interference components, significantly suppress stripe artifacts, scattering artifacts and boundary distortion artifacts generated during the imaging process, significantly improve the imaging clarity and edge contour integrity of the target tissue structure, effectively avoid the defects of traditional interpolation-based artifact removal algorithms that easily introduce secondary artifacts, loss of details and grayscale distortion, enhance the overall signal-to-noise ratio and anatomical detail recognition of the image, and ensure the accuracy and reliability of subsequent clinical diagnosis and image analysis.

[0032] In one example, the initial mask reconstruction image is segmented and projected to obtain a mask projection image with marked target regions. This includes: segmenting the initial mask reconstruction image based on a first threshold to mark the target regions of the initial mask reconstruction image; performing erosion and dilation processing on the initial mask reconstruction image with marked target regions to update the target regions in the initial mask reconstruction image, resulting in an eroded and dilated initial mask reconstruction image; and projecting the eroded and dilated initial mask reconstruction image based on a ray-driven model to obtain a mask projection image with marked target regions. The ray-driven model is implemented based on dual rays, where the rays are emitted when the device scans and acquires the image.

[0033] Specifically, an appropriate threshold (first threshold) is selected to segment the reconstructed sequence containing metal (initial mask reconstruction image). For the 3DDSA acquisition protocol, the metal region segmentation is performed on the first ring of mask slice sequence (initial mask reconstruction image). Since the threshold segmentation method will cause the intersection edge of soft tissue and metal to be too abrupt, the morphological erosion dilation method is used to process the metal edge position to obtain the eroded and dilated initial mask reconstruction image. The eroded and dilated initial mask reconstruction image is then forward-projected to obtain the coordinate position of the metal in the sinogram domain (projection domain). The forward projection map adopts a forward projection algorithm under the ray-driven model, which simulates the line integral by weighted summation of the pixels passed by the two rays.

[0034] In one example, the initial subtraction image is filtered based on the mask projection image to obtain an intermediate subtraction image. This includes: determining the afterimage region of the initial subtraction image based on the target region of the mask projection image; setting a second threshold; for pixels in the afterimage region, if the pixel gray value is greater than the second threshold and the number of pixels with gray values ​​greater than the second threshold in the connected component to which the pixel belongs is greater than the set threshold, the pixel gray value is retained; otherwise, the pixel gray value is set to a first preset value to obtain the intermediate subtraction image.

[0035] Specifically, due to the characteristics of small-area afterimages in the metal region (initial subtraction image) after subtraction in 3DDSA, non-connected regions are processed to prevent misprocessing of the vascular region, thereby obtaining a projection image (intermediate subtraction image) with metal afterimages removed.

[0036] For example, based on the position of the metal projection domain (mask projection image) obtained from the mask image, the position of the metal afterimage (afterimage region) in the subtracted projection domain (initial subtracted image) is obtained, and the gray value of the afterimage region is set to a non-vascular value (first preset value). At the same time, a vascular threshold (second threshold) and the number of connected pixels (set threshold) are set. The initial subtracted image is traversed to filter, thereby protecting the vascular position through connected components and obtaining an intermediate subtracted image.

[0037] In one example, image reconstruction processing is performed based on an initial mask reconstruction image and a mask projection image to determine a target mask reconstruction image. This includes: interpolating the target region in the mask projection image to obtain a first projection image; performing image reconstruction processing on the first projection image to obtain a first mask reconstruction image; binarizing the target region in the mask projection image to obtain a second projection image; performing image reconstruction processing on the second projection image to obtain a second mask reconstruction image; and determining the target mask reconstruction image based on the initial mask reconstruction image, the first mask reconstruction image, and the second mask reconstruction image.

[0038] Specifically, linear interpolation is performed on the metal region (target region) of the mask projection image. For example, the projection values ​​of the adjacent normal region (non-metal region) of the mask projection image are used. Based on the linear change law, linear fitting is performed according to the position distance weight. The interpolation calculates the alternative projection value of the metal region, realizing the smooth repair of the projection data contaminated by metal, thereby obtaining the first projection image. During interpolation, the value of the projection domain is limited to be greater than the threshold set in the projection domain to ensure that the interpolated value conforms to the physical attenuation law of X-rays and avoids the generation of negative values ​​and abnormally low values. The generation of metal artifacts is suppressed from the source of the projection domain, preventing secondary artifacts such as dark lines and black holes after reconstruction. This provides a reliable projection data foundation for subsequent high-quality image reconstruction and preservation of vascular information. The first projection image is then reconstructed to obtain the first mask reconstruction image (a slice sequence without metal regions and metal artifacts).

[0039] The target region in the mask projection image is binarized (e.g., thresholded binarization to 0 / 1). For example, a metal threshold (third threshold) is set for the mask projection image. The gray value of each pixel corresponding to the target region in the mask projection image is judged. When the gray value is greater than the third threshold, it is set to the second preset value (e.g., 1). When the gray value is less than or equal to the third threshold, it is set to the third preset value (e.g., 0). This yields the second projection image, which is used to count the number of scanning angles. The second projection image is then reconstructed to obtain the second mask reconstruction image (containing the metal region and metal artifacts, existing in the form of binary 0 / 1, and only used for subsequent counting of scanning angles).

[0040] The target mask reconstruction image is determined based on the initial mask reconstruction image, the first mask reconstruction image, and the second mask reconstruction image.

[0041] In the above embodiments, interpolation and binarization are performed on the target area of ​​the mask projection image to obtain a first projection image and a second projection image. The first mask reconstruction image, the target subtraction reconstruction image, and the second mask reconstruction image are then reconstructed by combining the intermediate subtraction image. Interpolation repair effectively compensates for abnormal metal data in the projection domain, reduces the interference of metal artifacts on the reconstructed image, and accurately restores the normal anatomical tissue structure. Preprocessing enables accurate identification of metal and artifact areas, providing a reliable basis for subsequent scanning angle statistics and pixel threshold determination. At the same time, the intermediate subtraction image is used to reconstruct a clean subtraction reconstruction image without metal afterimages or artifacts, improving the accuracy of 3DDSA metal artifact correction and the overall image quality.

[0042] In one example, the target region in the mask projection image is binarized to obtain a second projection image, including: judging the gray value of each pixel corresponding to the target region in the mask projection image based on a third threshold to obtain the second projection image, wherein when the gray value is greater than the third threshold, the gray value is set to a second preset value, and when the gray value is less than or equal to the third threshold, the gray value is set to a third preset value. Specifically, a metal threshold (third threshold) is set for the mask projection image, and the gray value of each pixel corresponding to the target area of ​​the mask projection image is judged. When the gray value is greater than the third threshold, it is set to a second preset value (e.g., 1). When the gray value is less than or equal to the third threshold, it is set to a third preset value (e.g., 0), thereby obtaining the second projection image.

[0043] In one example, the second mask reconstruction image includes multiple slice sequences; determining the target mask reconstruction image based on the initial mask reconstruction image, the first mask reconstruction image, and the second mask reconstruction image includes: obtaining a reference scanning angle for device scanning; determining the number of pixels with grayscale values ​​of a second preset value in each slice sequence of the second mask reconstruction image, and obtaining the scanning angle of each pixel based on the number; and determining the target mask reconstruction image based on the scanning angle, the reference scanning angle, the angle threshold, the initial mask reconstruction image, and the first mask reconstruction image.

[0044] Specifically, the reference scanning angle of the device is predetermined (e.g., 200°). Based on the scanning angle of each pixel in the second mask reconstruction image (including multiple slice sequences) (e.g., counting the number of pixels with grayscale values ​​of 1 in each slice sequence and using the sum of the numbers as the scanning angle of each pixel in the slice sequence), the target mask reconstruction image (slice sequence with metal regions and no metal artifacts) can be obtained by judging based on the scanning angle, reference scanning angle, angle threshold, initial mask reconstruction image (slice sequence with metal regions and no metal artifacts) and first mask reconstruction image (slice sequence without metal regions and metal artifacts).

[0045] In one example, determining the target mask reconstruction image based on the scanning angle, reference scanning angle, angle threshold, initial mask reconstruction image, and first mask reconstruction image includes: judging each pixel of the first mask reconstruction image based on the scanning angle, reference scanning angle, and angle threshold; when the scanning angle is greater than the angle threshold and less than the reference scanning angle, interpolating the first mask reconstruction image based on the initial mask reconstruction image to obtain the target mask reconstruction image corresponding to the pixel; when the scanning angle is equal to the reference scanning angle, using the gray value of the initial mask reconstruction image corresponding to the pixel as the gray value of the pixel in the first mask reconstruction image, and obtaining the target mask reconstruction image corresponding to the pixel based on the first mask reconstruction image.

[0046] Specifically, each pixel of the first mask reconstructed image is judged based on the scanning angle, reference scanning angle, and angle threshold as shown in formula (1): (1) Where slice_Image is the image with metal artifacts removed (target mask reconstruction image). Number of scanning angles (refer to scanning angles) The threshold value (angle threshold) for the scanning angle of the metal boundary is set. , The reconstructed image does not contain metal (first mask reconstructed image). Original reconstructed image (initial mask reconstructed image). The number of acquisition angles (scanning angles) that the metal area passed through.

[0047] When the scanning angle of a certain pixel is equal to the reference scanning angle ( The grayscale value of the pixel in the initial mask reconstruction image is used as the grayscale value of the pixel in the first mask reconstruction image, and the updated first mask reconstruction image is used as the target mask reconstruction image.

[0048] When the scanning angle is greater than the angle threshold and less than the reference scanning angle ( The first mask reconstruction image is interpolated based on the initial mask reconstruction image, and the interpolated first mask reconstruction image is used as the target mask reconstruction image.

[0049] In one example, interpolation processing is performed based on the initial mask reconstruction image and the first mask reconstruction image to obtain the target mask reconstruction image. This includes: when the scanning angle of a pixel is greater than an angle threshold and less than a reference scanning angle, determining a first difference between the reference scanning angle and the angle threshold, a second difference between the reference scanning angle and the scanning angle, and a third difference between the scanning angle and the angle threshold; determining a first ratio between the second difference and the first difference, and multiplying the gray value of the first mask reconstruction image corresponding to the pixel with the first ratio to obtain a first product result; determining a second ratio between the third difference and the first difference, and multiplying the gray value of the initial mask reconstruction image corresponding to the pixel with the second ratio to obtain a second product result; performing an addition operation based on the first product result and the second product result to obtain the gray value of the pixel in the first mask reconstruction image, and obtaining the target mask reconstruction image corresponding to the pixel based on the first mask reconstruction image.

[0050] Specifically, as shown in formula (1), when the scanning angle is greater than the angle threshold and less than the reference scanning angle ( ), determine the first difference The second difference i Third difference The first ratio of the second difference to the first difference , The grayscale value of each pixel in the first mask reconstructed image is obtained by calculating the first difference, the second difference, the third difference, the first ratio, and the second ratio. The target mask reconstructed image is then obtained based on the updated first mask reconstructed image.

[0051] In another example, artifact removal is performed directly in the image domain using a deep learning method. A residual network is used to learn the residual from the artifact image to the artifact, and then the predicted artifact is subtracted from the input artifact image containing the metallic artifact to obtain a clear image.

[0052] In another example, data repair and reconstruction based on the projection domain are performed using deep learning. First, the projected data regions contaminated by metal are located in either the projection or image domain. Then, the U-Net deep learning network is used to predict the projection values ​​of the metal-occluded areas. This is optimized within a unified deep learning framework by combining data consistency from the projection domain with prior knowledge from the image domain.

[0053] Figure 3 A flowchart of an image artifact removal method provided for another embodiment of this application is shown below. Figure 3 As shown, the image artifact removal methods include S301-S309.

[0054] S301, Input 3DDSA image.

[0055] For example, based on the 3DDSA acquisition protocol (a set of fixed parameters and procedures that the device must strictly follow when performing a three-dimensional rotational DSA to ensure that the angle, number of frames, dose, and contrast agent injection sequence are consistent for each scan), a 3DDSA device (a flat-panel C-arm digital subtraction angiography (DSA) with rotational scanning and three-dimensional reconstruction functions) is used to scan and obtain the initial mask image (mask image, obtained by rotational scanning without contrast agent injection), the initial subtraction image (sub image, obtained by digital subtraction based on the mask image and contrast image), and the initial contrast image (contrast image, obtained by scanning after contrast agent injection).

[0056] S302, Metal image segmentation based on mask image.

[0057] For example, the mask image is reconstructed to obtain an initial mask reconstruction image (including multiple slice sequences). The initial mask reconstruction image is then segmented. Through segmentation, soft tissue and metal regions can be marked. Furthermore, the boundary between the soft tissue and metal regions is optimized through erosion and dilation to obtain an optimized initial mask reconstruction image.

[0058] S303, forward projection yields the metal trajectory.

[0059] For example, the optimized initial mask reconstruction image is projected (forward projection) to obtain a mask projection image (metal trajectory) of the marked target region (metal region). The projection process is implemented based on a ray-driven model, which is based on a dual-ray model, where the rays are emitted by the 3DDSA device during image acquisition.

[0060] S304, removes metal afterimages from the sub image based on the metal trajectory.

[0061] For example, the initial subtraction image (sub image) is filtered based on the mask projection image to obtain the intermediate subtraction image, and the intermediate subtraction image is reconstructed.

[0062] S305, obtain the image of sub with metal afterimage removed.

[0063] For example, image reconstruction processing is performed on the intermediate subtraction image to obtain the target subtraction reconstructed image (sub image with metal afterimage removed) (e.g. Figure 4 As shown, the left side is the initial subtraction image containing metallic ghosting, and the right side is the target subtraction reconstructed image after removing metallic ghosting.

[0064] S306, Interpolate the metal positions in the original projection image of the mask according to the threshold, and preprocess the metal trajectory.

[0065] For example, linear interpolation is performed on the target region (metal location) of the mask projection image (original mask projection image). For instance, the projection values ​​of adjacent normal regions (non-metal regions) of the mask projection image are used. Based on the linear variation law, linear fitting is performed according to the positional distance weight to interpolate and calculate the alternative projection value of the metal region. This achieves smooth restoration of the metal-contaminated projection data, thereby obtaining the first projection image. The target region (metal trajectory) in the mask projection image is preprocessed (e.g., threshold binarization to 0 / 1). For example, a metal threshold (third threshold) is set for the mask projection image. The gray value of each pixel corresponding to the target region of the mask projection image is judged. When the gray value is greater than the third threshold, it is set to a second preset value (e.g., 1). When the gray value is less than or equal to the third threshold, it is set to a third preset value (e.g., 0), thereby obtaining the second projection image.

[0066] S307, Reconstruct the metal-free image and statistically reconstruct the scanning angles traversed by the reconstructed domain based on the preprocessed metal trajectory.

[0067] For example, the first projection image (metalless image) is reconstructed to obtain the first mask reconstruction image (a slice sequence of metalless regions and metal artifacts), and the second projection image is reconstructed to obtain the second mask reconstruction image (containing metal regions and metal artifacts, existing in the form of binary 0 / 1, used to count the scanning angle).

[0068] S308, obtain an image of the mask without metal artifacts.

[0069] For example, based on the initial mask reconstruction image, the first mask reconstruction image, and the second mask reconstruction image, a target mask reconstruction image (an image without metal artifacts on the mask) is obtained (e.g. Figure 5As shown, the left side is the initial mask reconstruction image containing metal artifacts, and the right side is the target mask reconstruction image with metal artifacts removed.

[0070] S309, Perform image fusion.

[0071] For example, the target subtraction reconstructed image (the image with metal artifacts removed) and the target mask reconstructed image (the image without metal artifacts masked) are fused to obtain the target image with artifacts removed.

[0072] In the above embodiments, the image artifact removal method can avoid artifact residue or correction failure caused by incorrect metal region segmentation, and can be applied to the removal of metal residue caused by 3DDSA patient movement and registration inaccuracy.

[0073] The image artifact removal method proposed in this application employs two constraints to control the error caused by inaccurate segmentation. First, the segmented metal region is eroded and expanded to obtain the most complete metal region possible (initial mask reconstruction image after erosion and expansion). Then, the metal region of the projection domain is located by forward projection (target region of the mask projection image). Based on the threshold of the metal projection domain set on the original projection image (mask projection image) of this region, the metal region is further accurately located. The metal region obtained above is eroded and expanded. Then, the angle of each point is counted in the reconstruction domain (second mask reconstruction image). When the counted angle is equal to the scanning angle, it is equal to the original metal value. When it is less than the scanning angle, the corresponding metal value is obtained by linear interpolation (target mask reconstruction image). Second, in the 3DDSA image, the metal region is segmented based on the first-ring reconstructed image mask (initial mask reconstructed image) to locate the position of the metal. The sub-image obtained by subtracting the mask image from the second-ring image is used to locate the area of ​​metal that has not been completely subtracted (the afterimage area) based on the position of the metal in the first-ring image (mask projection image). The residual metal area is removed by forward projection. After the mask metal artifact removal (target mask reconstructed image) and the sub metal afterimage removal (target subtraction reconstructed image) are merged, the information loss and blurring problems caused by directly processing on the contrast enhancement image full of artifacts are avoided.

[0074] Figure 6 A block diagram of an image artifact removal system provided for another embodiment of this application.

[0075] This specification provides an image artifact removal system 600. Please refer to [link / reference]. Figure 6 The image artifact removal system 600 includes: an acquisition module 610, a first processing module 620, a second processing module 630, a first reconstruction module 640, a second reconstruction module 650, and a fusion module 660.

[0076] The acquisition module 610 is used to acquire the initial mask image, the initial subtraction image, the initial contrast image, and the initial mask reconstruction image obtained by reconstructing the initial mask image, wherein the initial subtraction image is obtained by subtraction processing based on the initial mask image and the initial contrast image; The first processing module 620 is used to perform segmentation and projection processing on the initial mask reconstruction image to obtain a mask projection image of the marked target region. The second processing module 630 is used to filter the initial subtraction image based on the mask projection image to obtain the intermediate subtraction image; The first reconstruction module 640 is used to perform image reconstruction processing based on the intermediate subtraction image to obtain the target subtraction reconstructed image; The second reconstruction module 650 is used to perform image reconstruction processing based on the initial mask reconstruction image and the mask projection image to determine the target mask reconstruction image. The fusion module 660 is used to perform image fusion processing based on the target mask reconstruction image and the target subtraction reconstruction image to obtain the target image after artifact removal.

[0077] For example, the first processing module 620 is further configured to perform segmentation processing on the initial mask reconstruction image based on a first threshold to mark the target region of the initial mask reconstruction image; perform erosion and dilation processing on the initial mask reconstruction image with marked target regions to update the target regions in the initial mask reconstruction image to obtain the eroded and dilated initial mask reconstruction image; and perform projection processing on the eroded and dilated initial mask reconstruction image based on a ray-driven model to obtain a mask projection image with marked target regions, wherein the ray-driven model is implemented based on dual rays, and the rays are emitted when the device scans and acquires images. For example, the second processing module 630 is further configured to determine the afterimage region of the initial subtraction image based on the target region of the mask projection image; set the afterimage region to a first preset value; and perform filtering processing on the initial subtraction image based on a second threshold to obtain an intermediate subtraction image.

[0078] For example, the second reconstruction module 650 is further configured to perform interpolation processing on the target region in the mask projection image to obtain a first projection image, perform image reconstruction processing on the first projection image to obtain a first mask reconstruction image; perform binarization processing on the target region in the mask projection image to obtain a second projection image, perform image reconstruction processing on the second projection image to obtain a second mask reconstruction image; and determine the target mask reconstruction image based on the initial mask reconstruction image, the first mask reconstruction image, and the second mask reconstruction image. For example, the second reconstruction module 650 is further configured to determine the gray value of each pixel corresponding to the target region in the mask projection image based on the third threshold to obtain the second projection image, wherein when the gray value is greater than the third threshold, the gray value is set to the second preset value, and when the gray value is less than or equal to the third threshold, the gray value is set to the third preset value. For example, the second mask reconstruction image includes multiple slice sequences; the second reconstruction module 650 is further configured to obtain a reference scanning angle of the device scanning; determine the number of pixels with grayscale values ​​of a second preset value in each slice sequence of the second mask reconstruction image, and obtain the scanning angle of each pixel based on the number; and determine the target mask reconstruction image based on the scanning angle, the reference scanning angle, the angle threshold, the initial mask reconstruction image and the first mask reconstruction image. For example, the second reconstruction module 650 is further configured to determine each pixel of the first mask reconstruction image based on the scanning angle, the reference scanning angle, and the angle threshold. When the scanning angle is greater than the angle threshold and less than the reference scanning angle, the first mask reconstruction image is interpolated based on the initial mask reconstruction image to obtain the target mask reconstruction image corresponding to the pixel. When the scanning angle is equal to the reference scanning angle, the gray value of the initial mask reconstruction image corresponding to the pixel is used as the gray value of the pixel in the first mask reconstruction image, and the target mask reconstruction image corresponding to the pixel is obtained based on the first mask reconstruction image.

[0079] For example, the second reconstruction module 650 is further configured to: determine a first difference between the reference scanning angle and the angle threshold, a second difference between the reference scanning angle and the scanning angle, and a third difference between the scanning angle and the angle threshold when the scanning angle of the pixel is greater than an angle threshold and less than a reference scanning angle; determine a first ratio between the second difference and the first difference, and multiply the gray value of the first mask reconstruction image corresponding to the pixel with the first ratio to obtain a first product result; determine a second ratio between the third difference and the first difference, and multiply the gray value of the initial mask reconstruction image corresponding to the pixel with the second ratio to obtain a second product result; perform an addition operation based on the first product result and the second product result to obtain the gray value of the pixel in the first mask reconstruction image, and obtain the target mask reconstruction image corresponding to the pixel based on the first mask reconstruction image.

[0080] Figure 7 A block diagram of an electronic device provided for another embodiment of this application.

[0081] Another embodiment of this application provides an electronic device having a computer program stored thereon, which, when executed by a processor, implements the steps of the method of any of the above embodiments.

[0082] like Figure 7As shown, for ease of understanding, an embodiment of this application illustrates a specific electronic device 700.

[0083] Electronic device 700 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 700 may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0084] like Figure 7 As shown, the electronic device 700 includes a computing unit 701, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 708 into a random access memory (RAM) 703. The RAM 703 may also store various programs and data required for the operation of the electronic device 700. The computing unit 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0085] Multiple components in electronic device 700 are connected to input / output (I / O) interface 705. These components include: input unit 706, such as a keyboard or mouse; output unit 707, such as various types of displays or speakers; storage unit 708, such as a disk or optical disk; and communication unit 709, such as a network interface card (NIC), modem, or wireless transceiver. Communication unit 709 allows electronic device 700 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0086] The computing unit 701 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs the various methods described above. For example, in some embodiments, any one or more of the various methods described above can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 700 via ROM 702 and / or communication unit 709. When the computer program is loaded into RAM 703 and executed by the computing unit 701, one or more steps of any one or more of the various methods described above can be performed. Alternatively, in other embodiments, the computing unit 701 can be configured to perform any one or more of the various methods described above by any other suitable means (e.g., by means of firmware).

[0087] This application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method in any of the above embodiments.

[0088] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this application, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0089] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0090] In the description of this application, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this application, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0091] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0092] Furthermore, the terms "first," "second," etc., used in the embodiments of this application are for descriptive purposes only and should not be construed as indicating or implying relative importance, or implicitly specifying the number of technical features indicated in this embodiment. Therefore, features defined with terms such as "first" and "second" in the embodiments of this application can explicitly or implicitly indicate that the embodiment includes at least one of those features. In the description of this application, the word "multiple" means at least two or more, such as two, three, four, etc., unless otherwise explicitly and specifically defined in the embodiments.

[0093] In this application, unless otherwise explicitly specified or limited in the embodiments, the terms "installation," "connection," "joining," and "fixing" appearing in the embodiments should be interpreted broadly. For example, a connection can be a fixed connection, a detachable connection, or an integral part; it can also be a mechanical connection, an electrical connection, etc. Of course, it can also be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication between two components, or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific implementation.

[0094] In this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0095] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for removing image artifacts, characterized in that, The method includes: Acquire an initial mask image, an initial subtraction image, an initial contrast image, and an initial mask reconstruction image reconstructed from the initial mask image, wherein the initial subtraction image is obtained by subtraction processing based on the initial mask image and the initial contrast image; The initial mask reconstruction image is segmented and projected to obtain a mask projection image of the marked target region; Based on the mask projection image, the initial subtraction image is filtered to obtain an intermediate subtraction image; Image reconstruction processing is performed on the intermediate subtraction image to obtain the target subtraction reconstructed image; Image reconstruction processing is performed based on the initial mask reconstruction image and the mask projection image to determine the target mask reconstruction image; Image fusion processing is performed on the target mask reconstruction image and the target subtraction reconstruction image to obtain the target image after artifact removal.

2. The method according to claim 1, characterized in that, The step of segmenting and projecting the initial mask reconstruction image to obtain a mask projection image of the marked target region includes: The initial mask-reconstructed image is segmented based on a first threshold to mark the target region of the initial mask-reconstructed image; The initial mask reconstruction image of the marked target region is subjected to erosion and dilation processing to update the target region in the initial mask reconstruction image, resulting in an erosion and dilation-reconstructed initial mask reconstruction image; The initial mask reconstruction image after corrosion and expansion is projected based on the ray-driven model to obtain the mask projection image of the marked target area. The ray-driven model is based on dual rays, which are emitted when the device scans and acquires the image.

3. The method according to claim 1, characterized in that, The step of filtering the initial subtraction image based on the mask projection image to obtain an intermediate subtraction image includes: Based on the target region of the mask projection image, the afterimage region of the initial subtraction image is determined; A second threshold is set. For pixels in the afterimage region, if the pixel gray value is greater than the second threshold and the number of pixels in the connected component to which the pixel belongs that have a gray value greater than the second threshold is greater than the set threshold, the pixel gray value is retained. Otherwise, the pixel gray value is set to the first preset value to obtain the intermediate subtraction image.

4. The method according to claim 1, characterized in that, The step of performing image reconstruction processing based on the initial mask reconstruction image and the mask projection image to determine the target mask reconstruction image includes: The target region in the mask projection image is interpolated to obtain a first projection image, and the first projection image is reconstructed to obtain a first mask reconstruction image. The target region in the mask projection image is binarized to obtain a second projection image, and the second projection image is reconstructed to obtain a second mask reconstruction image. Based on the initial mask reconstruction image, the first mask reconstruction image, and the second mask reconstruction image, a target mask reconstruction image is determined.

5. The method according to claim 4, characterized in that, The step of binarizing the target region in the mask projection image to obtain a second projection image includes: Based on a third threshold, the grayscale value of each pixel corresponding to the target region in the mask projection image is determined to obtain a second projection image. When the grayscale value is greater than the third threshold, the grayscale value is set to a second preset value. When the grayscale value is less than or equal to the third threshold, the grayscale value is set to a third preset value.

6. The method according to claim 5, characterized in that, The second mask reconstruction image includes multiple slice sequences; determining the target mask reconstruction image based on the initial mask reconstruction image, the first mask reconstruction image, and the second mask reconstruction image includes: Obtain the reference scanning angle for the device; The number of times the grayscale value of a pixel in each slice sequence of the second mask-reconstructed image is determined to be the second preset value, and the scanning angle of each pixel is obtained based on the number. The target mask reconstruction image is determined based on the scanning angle, the reference scanning angle, the angle threshold, the initial mask reconstruction image, and the first mask reconstruction image.

7. The method according to claim 6, characterized in that, The step of determining the target mask reconstruction image based on the scanning angle, the reference scanning angle, the angle threshold, the initial mask reconstruction image, and the first mask reconstruction image includes: Based on the scanning angle, the reference scanning angle, and the angle threshold, each pixel of the first mask reconstruction image is judged. When the scanning angle is greater than the angle threshold and less than the reference scanning angle, the first mask reconstruction image is interpolated based on the initial mask reconstruction image to obtain the target mask reconstruction image corresponding to the pixel. When the scanning angle is equal to the reference scanning angle, the gray value of the initial mask reconstruction image corresponding to the pixel is used as the gray value of the pixel in the first mask reconstruction image, and the target mask reconstruction image corresponding to the pixel is obtained based on the first mask reconstruction image.

8. The method according to claim 7, characterized in that, The step of interpolating the first mask reconstruction image based on the initial mask reconstruction image to obtain the target mask reconstruction image includes: When the scanning angle of the pixel is greater than the angle threshold and less than the reference scanning angle, a first difference between the reference scanning angle and the angle threshold, a second difference between the reference scanning angle and the scanning angle, and a third difference between the scanning angle and the angle threshold are determined. Determine the first ratio between the second difference and the first difference, and multiply the gray value of the first mask reconstructed image corresponding to the pixel with the first ratio to obtain the first product result; Determine the second ratio between the third difference and the first difference, and multiply the gray value of the initial mask reconstructed image corresponding to the pixel with the second ratio to obtain the second product result; An addition operation is performed based on the first product result and the second product result to obtain the gray value of the pixel in the first mask reconstruction image, and the target mask reconstruction image corresponding to the pixel is obtained based on the first mask reconstruction image.

9. An image artifact removal system, characterized in that, The system includes: The acquisition module is used to acquire an initial mask image, an initial subtraction image, an initial contrast image, and an initial mask reconstruction image reconstructed from the initial mask image, wherein the initial subtraction image is obtained by subtraction processing based on the initial mask image and the initial contrast image; The first processing module is used to perform segmentation and projection processing on the initial mask reconstruction image to obtain a mask projection image of the marked target region. The second processing module is used to filter the initial subtraction image based on the mask projection image to obtain an intermediate subtraction image; The first reconstruction module is used to perform image reconstruction processing based on the intermediate subtraction image to obtain the target subtraction reconstructed image; The second reconstruction module is used to perform image reconstruction processing based on the initial mask reconstruction image and the mask projection image to determine the target mask reconstruction image; The fusion module is used to perform image fusion processing based on the target mask reconstruction image and the target subtraction reconstruction image to obtain the target image after artifact removal.

10. An electronic device having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1-8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1-8.

Images