A method for metal artifact correction of three-dimensional images

Abstract

The application discloses a kind of metal artifact correction methods of three-dimensional image, comprising: the original projection sequence image of metal implant is scanned and is carried out back projection reconstruction, and the metal region image and non-metal region image are obtained by segmenting after image reconstruction;Respectively, the projection image of corresponding region is obtained by carrying out forward projection to the metal region image and non-metal region image, and the pixel value of the metal region of the original projection sequence image is replaced according to the pixel value of non-metal region image, to obtain the projection sequence image without metal;The back projection reconstruction is carried out to the projection sequence image without metal, and the CT value of each point in the metal region is replaced by the CT value of the corresponding point in the metal region image, to obtain the image after metal artifact correction.The application can effectively avoid the problem of image quality reduction caused by interpolation error, improve the contrast of CT image and the clarity of tissue boundary.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a method for correcting metal artifacts in three-dimensional images. Background Technology

[0002] C-arm machines are increasingly being adopted by major hospitals due to their advantages such as low radiation dose and high spatial resolution. They are also increasingly used in the field of orthopedic surgical robots, providing three-dimensional image guidance for surgeons or surgical robots.

[0003] During use, it is essential to ensure that the C-arm provides clear and accurate 3D images. However, various image artifacts can be introduced due to various reasons, leading to degraded image quality and misinterpretations. Metal artifacts are a common type of image artifact. Since metal implants (such as metal stents, bone plates, and long screws) are often required during surgery, interference from these implants is unavoidable in the clinical application of CBCT. Currently, the most common metal correction methods are based on interpolation algorithms. However, when the metal implant is large, the correction value obtained by the interpolation algorithm will deviate significantly from its accurate value. Summary of the Invention

[0004] Purpose of the invention: To address the above-mentioned shortcomings, this invention proposes a method for correcting metal artifacts in 3D images without interpolation, thus avoiding the problem of image quality degradation caused by interpolation errors.

[0005] Technical solution:

[0006] A method for correcting metal artifacts in three-dimensional images, comprising:

[0007] The original projection sequence image of the metal implant was obtained by scanning and then back-projected to reconstruct it. The reconstructed image was segmented to obtain the metal region image and the non-metal region image.

[0008] The metal region image and the non-metal region image are forward-projected to obtain the corresponding region projection image. The pixel values ​​of the metal region in the original projection sequence image are replaced according to the pixel values ​​of the non-metal region image to obtain the metal-free projection sequence image.

[0009] The metal-free projection sequence image is back-projected and reconstructed, and the CT values ​​of each point in the metal region are replaced with the CT values ​​of the corresponding points in the metal region image to obtain the image after metal artifact correction.

[0010] The specific steps of replacing the pixel values ​​of the metal region image with the pixel values ​​of the non-metal region image to obtain the metal-free projection sequence image are as follows:

[0011] The non-metallic region in the original projection sequence image is divided by the pixel value of the projected image of the non-metallic region to obtain the corresponding average value. Based on this, the pixel value of each point in the metallic region in the original projection sequence image is replaced by the product of the average value and the pixel value of the corresponding point in the projected image of the metallic region, thereby obtaining a metal-free projection sequence image.

[0012] The pixel values ​​of each point in the metal region of the original projection sequence image are set to 0 based on the image projection of the metal region to obtain a preliminary projection sequence image. The pixel values ​​of the metal region in the preliminary projection sequence image are then replaced with the pixel values ​​of the non-metal region image to obtain a metal-free projection sequence image.

[0013] After obtaining the preliminary projection sequence image, median filtering is applied to it.

[0014] After obtaining the projected image of the non-metallic region, it is normalized.

[0015] The normalization process specifically involves:

[0016]

[0017] Where P(x,y,i) is the pixel value of the i-th projected image at point (x,y), P max and P min These represent the maximum and minimum pixel values ​​of all points in the projected image sequence obtained by forward projection, respectively.

[0018] P norm (x,y,i) represents the pixel value of the i-th normalized projected image at point (x,y).

[0019] Beneficial effects: This invention eliminates the need for interpolation, effectively avoiding image quality degradation caused by interpolation errors, and effectively improving the contrast and tissue boundary clarity of CT images. Attached Figure Description

[0020] Figure 1 This is a flowchart of the present invention;

[0021] Figure 2 A schematic diagram of a projection image containing a metal implant;

[0022] Figure 3 This is a schematic diagram of the projected image after replacing the location of the metal implant with air through threshold segmentation;

[0023] Figure 4 To pass Figure 2 A schematic diagram of a slice image obtained by reconstructing the projected image shown.

[0024] Figure 5 To pass Figure 3 The diagram shows a slice image after the projected image has been reconstructed and corrected. Detailed Implementation

[0025] The present invention will be further explained below with reference to the accompanying drawings and specific embodiments.

[0026] The method for correcting metal artifacts in three-dimensional images according to the present invention is as follows: Figure 1 As shown, it includes the following steps:

[0027] (1) Start the C-arm scanner, place the object with the metal implant in the center of the C-arm scanner's field of view, and perform a three-dimensional scanning operation to obtain the original projection sequence images, such as... Figure 2 As shown;

[0028] (2) Perform backprojection reconstruction on the original projection sequence image obtained in step (1) to obtain a reconstructed image without metal correction, such as... Figure 4 As shown;

[0029] (3) Threshold segmentation is performed on the reconstructed image obtained in step (2) to obtain the corresponding metal region image and non-metal region image;

[0030] (4) The non-metallic region image obtained in step (3) is forward-projected using the corresponding projection model to obtain the projection sequence image of the non-metallic region, and then normalized.

[0031] The normalization is as follows:

[0032]

[0033] Where P(x,y,i) is the pixel value of the i-th projected image at point (x,y), P max and P min P represents the maximum and minimum pixel values ​​of all points in the projected image sequence obtained by forward projection, respectively. norm (x,y,i) represents the pixel value of the i-th normalized projected image at point (x,y);

[0034] (5) The metal region image obtained in step (3) is forward-projected using a corresponding projection model to obtain a projected sequence image of the metal region. Based on this, the pixel values ​​of each point in the metal region in the original projected sequence image obtained in step (1) are set to 0 to obtain a preliminary projected sequence image. Median filtering is then applied to remove shot noise, such as... Figure 3 As shown;

[0035] (6) Perform a point-by-point division operation on the pixel values ​​of each point in the non-metallic region of the preliminary projection sequence image obtained in step (5) and the image obtained in step (4), and calculate the average value P of the corresponding pixel values ​​of each point. m ;

[0036] (7) Convert the pixel values ​​of each point in the metal region of the image obtained in step (5) using P m The pixel values ​​of the corresponding points in the projected sequence image of the metal region obtained in step (4) are replaced by the product of the pixel values ​​of the corresponding points in the metal region. The pixel values ​​of each point in the metal region are calculated accordingly, and the projected sequence image without metal is obtained accordingly.

[0037] (8) Perform backprojection reconstruction based on the metal-free projection sequence image obtained in step (7) to obtain the metal-free reconstructed image, such as... Figure 5 As shown;

[0038] (9) Replace the CT values ​​of each point in the metal region in the metal region image obtained in step (3) with the CT values ​​of each point in the metal region in step (8) to obtain the metal-suppressed image.

[0039] This invention addresses the problem of severe metal artifacts in reconstructed 3D CT images caused by metal implants, which degrades CT image quality and reduces the reliability of CT images as a basis for medical diagnosis. This invention avoids the tissue boundary blurring caused by deviations in the tissue attenuation coefficients between the interpolation point and the interpolated metal point, as well as interpolation errors, in conventional metal correction methods. It effectively improves the accuracy of CT values ​​in the interpolated metal point image, significantly enhancing CT image contrast and tissue boundary clarity.

[0040] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details of the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations (such as quantity, shape, position, etc.) can be made to the technical solution of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A method for correcting metal artifacts in three-dimensional images, characterized in that: include: The original projection sequence image of the metal implant was obtained by scanning and then back-projected to reconstruct it. The reconstructed image was segmented to obtain the metal region image and the non-metal region image. The metal region image and the non-metal region image are forward projected to obtain the corresponding region projection image. The non-metal region and the non-metal region projection image are divided by the pixel value of each point to calculate the corresponding average value. Based on this, the pixel value of each point in the metal region in the original projection sequence image is replaced by the product of the average value and the pixel value of the corresponding point in the metal region projection image, thereby obtaining the metal-free projection sequence image. The metal-free projection sequence image is back-projected and reconstructed, and the CT values ​​of each point in the metal region are replaced with the CT values ​​of the corresponding points in the metal region image to obtain the image after metal artifact correction.

2. The method for correcting metal artifacts in three-dimensional images according to claim 1, characterized in that: The pixel values ​​of each point in the metal region of the original projection sequence image are set to 0 based on the projection image of the metal region image to obtain a preliminary projection sequence image. The pixel values ​​of the metal region in the preliminary projection sequence image are then replaced with the pixel values ​​of the non-metal region image to obtain a metal-free projection sequence image.

3. The method for correcting metal artifacts in three-dimensional images according to claim 2, characterized in that: After obtaining the preliminary projection sequence image, median filtering is applied to it.

4. The method for correcting metal artifacts in three-dimensional images according to claim 1, characterized in that: After obtaining the projected image of the non-metallic region, it is normalized.

5. The method for correcting metal artifacts in three-dimensional images according to claim 4, characterized in that: The normalization process specifically involves: ； in, Let be the pixel value of the i-th projected image at point (x, y). These represent the maximum and minimum pixel values ​​of all points in the projected image sequence obtained by forward projection, respectively. Let be the pixel value of the i-th projected image at point (x, y) after normalization.

Images