JP2006000226A - X-ray ct apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To capture a tomogram with less artifact even when a highly absorbent article for X rays such as a metal exists. <P>SOLUTION: A reconstruction function which decreases a high-frequency element as shown in Figure 5 is used in a range R1 including the highly absorbent article M, and a standard reconstruction function such as Shepp & Logan function is used in a normal range R3. A reconstruction function which interpolates between the reconstruction function of the range of the highly absorbent article and the reconstruction function of the normal range corresponding to a distance from the range including the highly absorbent article M and gradually shifts is used in a shifting range R2. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a cone beam type X-ray CT apparatus for obtaining a tomographic image in which artifacts due to an X-ray superabsorber are reduced.

  In an X-ray CT (Computer Tomography) apparatus, if an X-ray high-absorber such as a metal is present inside a subject, a high-frequency component accompanying the attenuation of the X-ray intensity is included in the projection data. A streak artifact called a streak artifact is generated, which hinders diagnosis.

  In order to remove or reduce such artifacts, for example, in Patent Document 1, a projection image is first reconstructed to generate a tomographic image, and a region of interest related to metal or the like is set from the tomographic image. Then, a region corresponding to the region of interest on the projection data is obtained, and moving average processing is performed on this region. By reconstructing the moving average processed projection data, a tomographic image with reduced artifacts is obtained.

JP-A-6-98886

  The same reconstruction function is used when reconstructing a plurality of tomographic images for one patient in a cone beam X-ray CT apparatus or the like. For this reason, when a tomographic image is created by photographing a patient who has a metal inside and outside the body, the tomographic image of a cross section where the metal exists becomes an image unsuitable for diagnosis due to the occurrence of an artifact.

  Moreover, in the method of the above-mentioned patent document 1 for removing this artifact, a reconstruction process must first be performed in order to find the presence of a high-absorbent material such as metal or the existence region thereof from a tomographic image. Then, after correcting the projection data, the reconstruction process is performed again.

  That is, it is necessary to perform the reconstruction process twice. However, the reconstruction process is a process that requires an enormous amount of calculation and takes time. In particular, in the case of a cone beam CT apparatus using a two-dimensional flat panel detector as an X-ray detector, the slice pitch in the body axis direction can be reconstructed with the same fineness as the pixel pitch, and thus the reconstruction process is performed. Time will be tremendous.

  Further, if the moving average process is performed on the projection data in order to remove or reduce the artifact, the processing time increases, and further, the moving average process cannot perform an accurate correction.

  An object of the present invention is to provide an X-ray CT apparatus that solves the above-described problems, can maintain an imaging cycle without loss of calculation time, and is effective for efficient imaging and diagnosis.

  In order to achieve the above object, an X-ray CT apparatus according to the present invention includes a determination unit that extracts a superabsorber region from a projection image and determines whether or not a cross section to be reconstructed includes the high absorber, and the high absorption A first reconstructing means for reconstructing a cross section not including a body using a normal first reconstruction function; and a second high artifact reducing effect for cutting a higher frequency with respect to the cross section including the high absorbent. A second reconstructing means for reconstructing using a reconstruction function, and a cross section in the vicinity of the cross section including the superabsorber, from the second reconstruction function to the first reconstruction function according to the distance thereof. And third reconstruction means for reconstructing using a third reconstruction function that gradually shifts.

  Further, the X-ray CT apparatus according to the present invention obtains a distance between the superabsorber region extracted from the projection image and a point to be reconstructed, and a range for obtaining data to be back-projected based on the distance. And a reconstruction means for reconstructing by reprojecting data obtained by interpolating projection data within the range.

  According to the X-ray CT apparatus of the present invention, it is possible to obtain a tomographic image with less metal artifacts by changing the reconstruction function, which is effective for diagnosis and only changes the reconstruction function. The imaging cycle can be maintained without loss of time, and efficient imaging and diagnosis can be performed effectively.

  Further, by making the data range to be backprojected variable according to the high absorber, a natural tomographic image with few artifacts can be obtained.

  The present invention will be described in detail based on the embodiments shown in the drawings.

  FIG. 1 is a block diagram of an X-ray CT system according to the present invention. An X-ray source 1 and an X-ray detector 2 are disposed across a subject S as a patient, and an output of an imaging system control unit 3 is connected to the X-ray source 1 via an X-ray generation control unit 4. . The output of the X-ray detector 2 is connected to the imaging system control unit 3 via the image input unit 5. Further, an image processing unit 6, an image storage unit 7, a diagnostic monitor 8, an operation unit 9, and a network 10 are connected to the imaging system control unit 3, and a printer 11, a diagnostic workstation 12, and an image database are connected to the network 10. 13 is connected.

  X-rays generated from the X-ray source 1 controlled by the X-ray generator control unit 4 pass through the subject S as a patient and are detected by the X-ray detector 2, and the detected X-rays are imaged as projection images. Input to the input unit 5. The X-ray source 1 and the X-ray detector 2 collect projection images at every predetermined rotation angle while rotating around the subject S as a rotation center. Alternatively, the subject S fixed to the rotary table may be rotated while maintaining the positional relationship between the X-ray source 1 and the X-ray detector 2.

  The input projection image at each rotation angle is subjected to image processing such as preprocessing and reconstruction processing including correction and logarithmic conversion of the X-ray detector 2 by the image processing unit 6 via the imaging system control unit 3. A tomographic image group is created. The created tomographic image group is displayed on the diagnostic monitor 8, stored in the image storage unit 7, and output to the printer 11, the diagnostic workstation 12, and the image database 13 via the network 10. Various operations such as a display window operation, a tomographic image switching display operation in the body axis direction, a cross-section conversion operation, and a three-dimensional surface display operation are performed by the operation unit 9.

  FIG. 2 is a flowchart of CT imaging in this system. First, in step S201, imaging conditions such as tube voltage, tube current, and exposure time, patient information such as patient name, age, and sex, and examination information such as an examination ID are set. When shooting is performed in step S202 in accordance with the set shooting conditions, pre-processing is performed in step S203 on the projection image obtained by shooting.

  Then, the tomographic image is created by reconstructing the preprocessed projection image in step S204. The created tomographic image is displayed in step S205, and the display image is confirmed in step S206 to determine the success or failure of imaging.

  If imaging has failed, the process returns to step S202 to perform re-imaging. If imaging has succeeded, the tomographic image created in step S207 is transferred to an image server or the like, and the process proceeds to the next imaging.

  FIG. 3 shows a flowchart of the process. First, the superabsorbent is extracted from the projection image obtained by photographing in step S301. Since the X-ray attenuation is large in the high absorber, the pixel value of the high absorber region is a relatively small value on the projection image.

  Therefore, binarization is performed by threshold processing, and an approximate shape is extracted. By repeating expansion processing and erosion processing, which are binary image processing, extra small components and holes can be removed. Therefore, it is possible to extract the high-absorber region clearly without waste. When the superabsorber region is extracted, the slice position corresponding to the superabsorber region extracted in step S302 is identified.

  As shown in FIG. 4, since the vertical direction of the projection image P corresponds to the slice direction, the vertical range in which the high-absorber M such as metal in the subject S exists is identified, and the high-absorber M is identified. Is the slice position range in which. In step S303, a transition range is set above and below the slice position range. As the width of this range, a width obtained by expanding or reducing the width of the slice position range at a predetermined rate may be used.

  In this way, since the slice position range, transition range, and other normal ranges in which the superabsorbent M exists are determined in step S304, a reconstruction function at the time of reconstruction is set within each range.

  As shown in FIG. 4, the reconstruction function uses a reconstruction function that reduces the high-frequency component as shown in FIG. 5 in the range R1 in which the high-absorber M is included, and a standard such as the Shepp & Logan function in the normal range R3. A general reconstruction function is used. In the transition range R2, a reconstruction function that gradually transitions by interpolating the reconstruction function of the high absorber range and the reconstruction function of the normal range according to the distance from the range including the high absorber M is used.

  When the reconstruction function set in this way is reconstructed in step S204 of FIG. 2, the reconstruction function is selected and reconstructed according to the slice position.

  In this manner, a normal reconstruction image is obtained for a cross section that does not include the superabsorbent body M, and a reconstructed image with reduced artifacts is obtained for a cross section that includes the superabsorbent body M. And in the cross section of the transition area | region from the cross section containing the high absorber M to the cross section which does not contain the high absorber M, the change between cross sections can be made smooth.

  In the second embodiment, similarly to the first embodiment, the pre-processing in step S203 and the reconfiguration processing in step S204 in the flowchart of FIG. 2 are executed. FIG. 6 shows a flowchart of the second embodiment in this part.

  Here, as in the first embodiment, the superabsorber is extracted from the projection image obtained by photographing. In the reconstruction process, after the convolution process, corresponding data in all projected images is back-projected for each reconstruction point. Steps S602 to S604 are executed in order to adaptively calculate the corresponding data in accordance with the distance from the high absorber. First, in step S602, the distance between the reconstruction point on the projection image and the superabsorber is obtained.

  For this purpose, the coordinates of the reconstruction point on the projection image are obtained. As shown in FIG. 7, the coordinate system is the three-dimensional absolute coordinate system (X, Y, Z), the tomographic plane to be reconfigured is the (X, Y) plane of the three-dimensional absolute coordinate system, and the slice direction is the Z axis. Take it.

Assuming that the rotation center of the X-ray source 1 is the origin O (0, 0, 0) of the three-dimensional absolute coordinate system, the coordinates of the rotation start position of the X-ray source 1 are (−D0, 0, 0). To do. The coordinates of the position of the X-ray source 1 when the X-ray source 1 is rotated by the angle θ are (−D0 cos θ, −D0 sin θ, 0). Then, assuming that the reconstruction point is (r _x , r _y , r _Z ), the projection plane passes through the origin O from the X-ray source 1 and is perpendicular to the position D0 + D1 on the opposite side, and O ′ on the projection plane. A local two-dimensional coordinate system (x, y) with the origin as.

If the coordinates of the reconstructed point in the projection plane and the local two-dimensional coordinate system are (Px, Py), they can be expressed as follows.
_{Px = {r x -t × (} D0 cosθ + r x) -D1 cos} × cosθ- {r y -t × (D0 sinθ + r y) -D1 sinθ} × sinθ
Py = r _z −t
_{However, t = {(r x -D1} cosθ) × D1 cosθ + (r y -D1 sinθ) × D1 sinθ} / {(r x -D0 cosθ) × D1 cosθ + (r y -D0 sinθ) × D1 sinθ}

  The distance between the reconstruction point (Px, Py) and the superabsorber is obtained. As described above, the superabsorber region can be obtained as a binary image, and the boundary line of the superabsorber region is obtained. As a method for extracting the boundary line, for example, an image processing algorithm described in Non-Patent Document 1 may be used. Thereby, the coordinates Qn (Xn, Yn) of the point sequence of the boundary line of the high absorber are obtained.

Latest trends in image processing algorithms: Mikio Takagi et al., New Technology Communications Inc. 5. Geometric feature processing (2)

The distance PQ between the reconstructed point and the superabsorber is obtained from the minimum distance between the reconstructed point and the boundary point sequence. That is, PQ = min√ {(Px−Xn) ² + (Py−Yn) ² ).

  In this way, when the distance between the reconstruction point and the superabsorber is obtained in step S602 in FIG. 6, the backprojection data range corresponding to this distance is determined in step S603. For example, the table shown in Table 1 below may be used in which the distance defined by the pixel size and the backprojection data range are represented by the number of horizontal pixels × the number of vertical pixels.

Table 1
Distance range
0 to 1 7 × 7
1-2 6 × 6
2-3 3 × 5
3-4 4x4
4-5 3 × 3

  Also, the number of horizontal pixels and the number of vertical pixels in the backprojection data range can be made different sizes. For example, as shown in FIG. 8, if a predetermined number of points centered on a part of the boundary line of the superabsorber closest to the reconstruction point, for example, the shortest distance point, is vertically long, Make the projection data range horizontally long. On the contrary, if the point sequence is horizontally long, the back projection data range is vertically long as shown in FIG.

  When the back projection range is obtained, back projection data is obtained by interpolating projection image data within this range in step S604. For the interpolation, an interpolation function such as a spline function may be used, or data obtained by adding weights obtained by normalizing the reciprocal of the distance from the reconstruction point may be used.

  By reprojecting the backprojection data obtained in this way, a reconstructed image reduced by artifacts can be created.

  A method of reducing artifacts in back projection processing in reconstruction processing after extracting a high-absorber region from projection image data will be described. The most commonly used method as reconstruction processing of an X-ray CT apparatus Is a convolution (calculation) / back projection (back projection calculation) method.

In the cone beam CT, the Feldkamp method, which is an extension of this convolution back projection method to a cone beam, is used. This Feldkamp method can be expressed as image = BP [w ₁ × {w ₂ = proj * func}] Λ.

Here, image is a reconstructed image, proj is projection data, func is convolution function data called a reconstruction function, w ₁ and w ₂ represent weights, * represents convolution, and BP represents back projection. Yes.

  FIG. 10 is an explanatory diagram of a conventional backprojection. In the backprojection, the data of a point on the projected image P of the X-ray path from the X-ray source 1 passing through the reconstruction point A is made one round (half round + fan in the half scan). This is done by adding everything over the angle. However, since the points on the projected image P of the X-ray path do not necessarily coincide with the center point of the pixel of the projected image P, data obtained by interpolating four points in the vicinity of the point is added as backprojected data. .

  In the present invention, the data used for this backprojection are not all four points in the vicinity, but depending on the distance between the reconstruction point A and the extracted superabsorbent M on the projection image P as shown in FIG. The range is variable. That is, if the distance to the high absorber M is short, data over a wider range is used by interpolation, and if the distance to the high absorber M is long, the range is narrowed. Thereby, the influence of the high frequency component in the vicinity of the high absorber M can be reduced, and a reconstructed image with reduced artifacts can be obtained.

1 is a configuration diagram of an X-ray CT system. It is a flowchart figure of the flow of imaging | photography. It is a flowchart figure of the flow of a process. It is explanatory drawing which changes a reconstruction function with a cross-sectional position. It is a graph of the frequency characteristic of a reconstruction function. It is a flowchart figure of the flow of a process. It is explanatory drawing of the geometric positional relationship of a reconstruction point and a projection surface. It is explanatory drawing in case the partial line segment of the boundary line of the superabsorber nearest the reconstruction point is vertically long. It is explanatory drawing when the partial line segment of the boundary line of the superabsorber nearest to a reconstruction point is horizontally long. It is explanatory drawing of the conventional back projection. It is explanatory drawing of the variable backprojection data range.

Explanation of symbols

1 X-ray source 2 X-ray detector 3 Imaging system control unit 4 X-ray generation control unit 5 Image input unit 6 Image processing unit 7 Image storage unit 8 Diagnostic monitor 9 Operation unit

Claims (4)

  Determining means for determining whether or not a cross section to be reconstructed by extracting a superabsorber region from the projection image includes a high absorber, and using a normal first reconstruction function for the cross section not including the superabsorber; First reconstructing means for reconstructing, and second reconstructing means for reconstructing using a second reconstruction function having a high artifact mitigation effect that cuts higher frequencies for a cross section including the superabsorbent body, A third reconstruction is performed using a third reconstruction function that gradually shifts from the second reconstruction function to the first reconstruction function according to the distance of a section in the vicinity of the section including the superabsorbent. An X-ray CT apparatus comprising:   Determining means for obtaining a distance between the superabsorber region obtained by extracting the superabsorber region from the projection image and a point to be reconstructed, and determining a range for obtaining data to be backprojected according to the distance; and a projection within the range An X-ray CT apparatus comprising: reconstruction means for performing reconstruction by back projecting data obtained by interpolating data.   The determining means determines the backprojection data range from the distance between the superabsorber region reconstruction points in advance using a table of the distance between the superabsorber region reconstruction points and the backprojection data range. The X-ray CT apparatus according to claim 2.   If the partial line segment of the boundary line of the superabsorber nearest to the reconstruction point is vertically long, the backprojection data range is horizontally long.If the partial line segment is horizontally long, the backprojection data range is vertically long. The X-ray CT apparatus according to claim 3, wherein

Images