Method for reducing metal artifact in computed tomography
A method for reducing metal artifacts in computed tomography (CT) is disclosed. The method for reducing metal artifacts in CT includes: obtaining a sinogram reduced in size from an original sinogram; setting up a linear algebraic equation according to remaining data excluding data damaged by a metal based on the obtained sinogram; restoring a low-resolution image based on the set up linear algebraic equation; calculating a sinogram from the restored low-resolution image; restoring a sinogram by disposing the calculated sinogram data in the original sinogram and by utilizing the calculated sinogram data as pre-information; and restoring a final CT image from the restored sinogram. Through introduction of a novel metal artifact reduction (MAR) technique referred to as an algebraic correction technique (ACT) using an intermediate image of an attenuation coefficient of an outside of a metal area, an image closest to an original image can be obtained by minimizing metal artifacts in CT.
1. A method for reducing metal artifacts in computed tomography (CT), comprising:
a) obtaining a sinogram reduced in size from an original sinogram;
b) setting up a linear algebraic equation according to remaining data excluding data damaged by a metal based on the obtained sinogram;
c) restoring a low-resolution image based on the set up linear algebraic equation;
d) calculating a sinogram from the restored low-resolution image;
e) restoring a sinogram by disposing the calculated sinogram data in the original sinogram and by utilizing the calculated sinogram data as pre-information; and
f) restoring a final CT image from the restored sinogram.
2. The method according to claim 1 , wherein, in operation a), the original sinogram is reduced in size in a direction in which a detector is placed.
3. The method according to claim 1 , wherein, in operation b), the linear algebraic equation is represented by the following equation:
∑
j
=
1
N
w
ij
f
j
=
p
i
,
i
=
1
,
2
,
…
,
M
where M is the number of projection measurement objects; N is the number of grids; w ij is a weighting coefficient equal to a fragmentary length or area of a j-th image pixel intercepted by an i-th ray; and p i is measured projection data having M pixels.
4. The method according to claim 1 , wherein, in operation c), the low-resolution image is restored using Tikhonov regularization.
5. The method according to claim 1 , further comprising: calculating a metal area after restoration is performed in operation e).
6. The method according to claim 5 , further comprising: performing interpolation by reflecting data for the calculated metal area in restoring the final CT image in operation f).