X-ray or gamma ray systems or devices – Specific application – Computerized tomography
Reexamination Certificate
2001-04-05
2002-07-30
Bruce, David V. (Department: 2882)
X-ray or gamma ray systems or devices
Specific application
Computerized tomography
C378S008000, C378S901000
Reexamination Certificate
active
06426988
ABSTRACT:
BACKGROUND OF THE INVENTION AND RELATED ART STATEMENT
The invention relates to an image processing method of an X-ray CT, an X-ray CT, and an X-ray CT image-taking recording medium, especially, to a technique for reducing a false image or artifact by absorption or dispersion of X-rays by an X-ray high-absorber, such as metal.
Generally, as shown in FIG.
8
(
a
), an X-ray CT includes an X-ray tube
51
for generating an X-ray beam B and an X-ray detector
52
for detecting X-rays. The X-ray tube
51
and the X-ray detector
52
are disposed to sandwich an object P to be tested, and the X-ray tube
51
irradiates the X-ray beam B to the object P to be tested to take images while rotating around an axis Z of the object P (in a direction extending vertically with respect to a sheet surface in FIG.
8
(
a
)). Generally, the X-ray detector
52
includes a plurality of detecting elements CHi (i=1, 2 . . . , n−1, n wherein n is a natural number), and the detecting elements CHi are disposed in a fan shape every fine angle. (FIG.
8
(
a
))
In a conventional image processing method, X-rays are irradiated from the circumference of the object P to be tested to obtain X-ray transmission data f(P) on a projection plane U (FIG.
8
(
b
)). Incidentally, in the present specification, the X-ray irradiation is referred to as “the object P to be tested is projected in an irradiation direction of the X-ray beam B” (hereinafter, also abbreviated to “the object P to be tested is projected”). The X-ray transmission data f(P) obtained on the projection plane U at this time is also referred to as “measured projection data f(P)”. Further, the measured projection data f(P) on the projection plane F is subjected to a reconstruction process, such as filtering and backprojection, to obtain an original image G[f(P)] on an image plane V (FIG.
8
(
c
)). In the present specification, the filtering is defined such that a convolution integral calculus (superposition integral calculus) is carried out by using a convolution kernel function. Also, the method where the projection data is subjected to the reconstruction process, such as filtering and backprojection, is generally known as a filtered backprojection (hereinafter referred to as FBP) method. Incidentally, since the FBP method is one of representative reconstruction processing methods, explanation thereof is omitted. By the way, derivation of the projection data of an image by using a mathematical algorithm is called as “forward projecting the image in an irradiation direction of X-rays” (hereinafter, if applicable, abbreviated as “forward projecting the image”).
Incidentally, a row of the detecting elements CH
i
and a row of angles &thgr; in an irradiation direction of X-ray beam B are shown in a horizontal axial direction and in a vertical axial direction on the projection plane U, respectively. Also, hatching areas in the projection plane U and the image plane V are shown illustratively. Further, the projection plane U and image plane V show areas. Based on the above, the following explanation is provided.
Incidentally, in case a certain object is represented by “P”, the projection of the object P is indicated by f(P); in case a fault image is represented by “&agr;”, the forward projection of the fault image &agr; is indicated by F(&agr;); and in case certain projection data is represented by “&bgr;”, the fault image obtained by filtered and backprojection of the projection data &bgr;, i.e. the image reconstituted by the FBP method is indicated by B[H(&bgr;)] or G(&bgr;). At this time, it is assumed that an operation of the backprojection is represented by “B”; an operation of the filtering is represented by “H”; and an operator of the backprojection and filtering is represented by “G”. Also, in order to make distinction between the projection of the object to be tested and the forward projection of the image, the symbols with respect to the projection and the forward projection are represented by “f” and “F”. Further, since there is a measurement error, in case the projection data &bgr; is measured, even if the fault image obtained by the FBP method of the measured projection data &bgr; is forward projected, the forward projected fault image does not return to exactly the same data as the originally measured projection data &bgr;. Therefore, in the present specification, assuming that &bgr;=F[G(&bgr;)] does not hold, the following explanation is made. Incidentally, the projection data (including also the measured projection data f(P)) and the image (including also the measured fault image G[f(P)]) have numeral weights, different from the projection plane F and the fault plane G, in other words, they are dealt as pixel values in the following explanations of the present specification.
However, in case of the conventional image process method, there are the following problems.
In detail, when a reconstruction of the image is carried out by taking an image of an object to be tested, false images are generated. Especially, in case of taking an image of an object to be tested including a high-absorber consisting of a metal or the like, the false images generated at and around the high-absorber become conspicuous due to absorption or dispersion by the high-absorber. Hereinafter, in the present specification, the above-explained false image is called as “artifact”. In the false images, especially, there are known streak artifacts where radially striped patterns are generated around the high-absorber, and shading artifacts which are generated at portions sandwiched by a plurality of high-absorbers. In order to reduce the artifacts including the above-mentioned artifacts, various methods have been proposed. As representative reducing methods, there are mentioned an iterative reconstruction and reprojection (hereinafter referred to as “IRR”) method, and algebraic reconstruction technique/expectation and maximization (hereinafter referred to as “ART/EM”) method.
First, the IRR method is explained with reference to FIG.
9
. In the IRR method, an original fault image G[f(P)] on the fault plane V obtained by a conventional image processing method is again forward projected by a mathematical algorithm in an irradiation direction of X-rays to obtain forward projection data F(G[f(P)]) on the projection plane U. Then, an operator sets a portion corresponding to a high-absorber as a high-absorber area M on the image plane V with reference to the original fault image G[f(P)] on the fault plane G ((a) in FIG.
9
). Incidentally, the high-absorber area M is a closed area. A portion L where X-rays pass through the high-absorber area is formed in the projection plane U. A pixel value of the measured projection data f(P) with respect to the portion L where X-rays pass through the high-absorber area is replaced by a pixel value of the forward projection data F(G[f(P)]) or a pixel value derived from the forward projection data F(G[f(P)]) to correct the measured projection data f(P) and obtain a corrected projection data F(P
1
). It should be noted that the forward projection data F(G[f(P)]) is different from the corrected projection data F(P
1
) ((b) in
FIG. 9
) Further, reconstruction of the corrected projection data F(P
1
) is carried out again to obtain a corrected fault image G[F(P
1
)] on the image plane V. ((c) in
FIG. 9
)
Incidentally, “the pixel value of the measured projection data f(P) of the portion L where X-rays pass through the high-absorber area is replaced by the pixel value of correcting forward projection data F(G[f(P)])” means that the corrected projection data F(P
1
) is obtained by following equations (1) and (2).
In the portion where X-rays do not pass through the high-absorber area:
F
(
P
1
)=
f
(
P
) (1)
In the portion L where X-rays pass through the high-absorber area:
F
(
P
1
)=
F
(
G[f
(
P
)]) (2)
Incidentally, “the pixel value of the measured projection data f(P)
Ueno Yoshihiro
Yamada Yosihiro
Bruce David V.
Kanesaka & Takeuchi
Shimadzu Corporation
LandOfFree
Image processing method of X-ray CT, X-ray CT and X-ray CT... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image processing method of X-ray CT, X-ray CT and X-ray CT..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processing method of X-ray CT, X-ray CT and X-ray CT... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2826847