JP2000333944A - X-ray ct system and forming method of fluoroscopic image therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To quickly form a fluoroscopic image by arranging an enlargement operation device for calculating the enlargement ratio of respective images when projecting a cutting image obtained by cutting a three-dimensional image and an accumulative operation device for accumulatively arithmetically operating projection data of the cut image projected on it with every pixel of the fluoroscopic image. SOLUTION: A plurality of slice surfaces are scanned on an attention part of an examine by using a scan device 61, and a reconstitutional operation 62 of slice images of the slice surfaces of the examinee is performed with every single scan on the basis of a series of obtained scan data. Next, required image data between the slice images is determined by a data interpolating operation 64 to make image data on a three- dimensional image of the attention part. Next, a visual point and a plane of projection are set in forming a fluoroscopic image to calculate the enlargement ratio 66 of the respective images when projecting a cutting image obtained by cutting the three- dimensional image. An accumulative operation 67 is performed on projection data on the cutting image projected on it with every pixel of the fluoroscopic image to make image data on the fluoroscopic image.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method using an X-ray CT apparatus, and more particularly to a technique for creating a fluoroscopic image using a series of slice images acquired by the X-ray CT apparatus.

[0002]

2. Description of the Related Art In an X-ray CT apparatus, an X-ray source and a multi-channel X-ray detector are rotated around a body axis of a subject while facing each other, and X-ray attenuation data ( Hereinafter, X-ray measurement data) is collected, and the X-ray measurement data is subjected to image processing to reconstruct an image of a cross section (hereinafter, referred to as a slice image). Usually, collection of X-ray measurement data is performed over a plurality of cross sections in the body axis direction of the subject,
A plurality of slice images are reconstructed at a time. Also,
The image data of the series of slice images is replaced with three-dimensional image data, and a three-dimensional image of the target region of the subject is created.

[0003] Conventionally, when a fluoroscopic image of a region of interest of a subject is created using an X-ray CT apparatus, image data of the above-described three-dimensional image (hereinafter referred to as three-dimensional image data) is used. Here, the fluoroscopic image is equivalent to an image obtained when an attention site of the subject is imaged under a fluoroscopic condition using a normal diagnostic X-ray apparatus, and a three-dimensional image of the attention site is obtained.
A three-dimensional image is projected as a two-dimensional image on one plane.

In creating a perspective image, it is necessary to determine a viewpoint corresponding to the focal point of the X-ray tube and a projection plane on which the three-dimensional image data is projected, in addition to the three-dimensional image data. When the three-dimensional image data of the attention site (corresponding to the perspective field), the viewpoint of the perspective, and the projection plane are determined, the perspective image is obtained by the following procedure. First, a plurality of paths of X-rays projected from a perspective point of view, through a site of interest, and projected onto a perspective projection plane are selected. The number of X-ray paths is determined in relation to the resolution of the fluoroscopic image. Next, projection data is calculated for each X-ray path. In this calculation, the interpolation value data from the three-dimensional image data on the X-ray path or the three-dimensional image data in the vicinity of the X-ray path is obtained for the entire pass band in which each X-ray path passes through the site of interest, and the image data is obtained. And the accumulated value is obtained as projection data of the perspective image.

[0005]

However, in the above-described method for creating a perspective image, (1) the calculation for obtaining the X-ray path is complicated, and (2) the calculation time is too long because the number of calculation points is large. That, there was a problem.

In view of the above, the present invention provides an X-ray CT apparatus and a method for generating a fluoroscopic image of a fluoroscopic image creating method using an X-ray CT apparatus, which can simplify the operation procedure and speed up the arithmetic processing. The purpose is to:

[0007]

In order to achieve the above object, an X-ray CT apparatus according to the present invention comprises: a scanner having an X-ray source and an X-ray detector opposed to each other; In an X-ray CT apparatus including an image reconstruction device for reconstructing a plurality of slice images of a subject based on data, an image storage device for collectively collecting and accumulating a series of slice image data and a shortage of a series of slice images are insufficient. Interpolation calculation device that creates image data of a three-dimensional image by interpolating and calculating image data, data for setting the position of the viewpoint and projection plane for projecting the three-dimensional image, and creation on the projection plane A position input device for inputting data for setting the structural specifications of a perspective image to be projected, and calculating a projection magnification rate of each image when projecting a section image obtained by cutting a three-dimensional image, and calculating each pixel. The size of An enlargement operation device for performing enlargement operation and projection coordinate calculation of the image, and an accumulation operation device for accumulating operation of projection data of a section image projected on each pixel of the perspective image to create image data of the perspective image And an image display device for displaying a perspective image (claim 1).

The X-ray CT apparatus having this configuration is characterized by a position input device, an enlargement operation device, and an accumulation operation device. According to these devices, steps (5) to (10) of claim 2 are performed.
Is executed. That is, the viewpoint input, the projection plane, and the structural specifications of the perspective image on the projection plane (the number of pixels, the size of the pixels, the arrangement of the pixels, the reference point for projection, etc.) when the perspective image is created by the position input device. ) Is set, and the enlargement calculation device cuts the three-dimensional image, calculates the projection enlargement ratio of each pixel of the cross-sectional image, enlarges the pixel, and calculates the coordinates of the pixel of the cross-sectional image and the coordinates of the pixel of the perspective image. Are calculated by the accumulator, and the accumulating device performs an accumulative operation on the projection data of the cross-sectional image on the pixels of the fluoroscopic image to create image data of the fluoroscopic image.

In order to achieve the above object, a method for creating a fluoroscopic image according to the present invention is to scan a subject with an X-ray CT apparatus and to reconstruct an arbitrary image based on a series of slice images obtained by image reconstruction. (1) X-ray CT
(2) scanning the target region of the subject using the apparatus and collecting scan data; and (2) reconstructing a slice image of a plurality of slices based on the scan data collected in step (1). 3) a step of collectively collecting and accumulating the image data of the slice images obtained in step (2) as a series of slice image data; and (4) finding missing data by interpolation based on the series of slice image data. (3) a step of creating a three-dimensional image of the region of interest; (5) a step of setting a viewpoint position and a projection plane for projecting the three-dimensional image;
A step of cutting the three-dimensional image in parallel with a plane facing the viewpoint and setting a cut plane of a plurality of layers; and (7) a step of calculating a projection magnification of each pixel of the cut plane image.
(8) a step of setting the structural specifications of the perspective image created on the projection plane; (9) a step of projecting image data on each pixel of the cut image onto the pixels of the perspective image; and (10).
(1) accumulating projection data of a cross-sectional image for each pixel of a perspective image to create image data of the perspective image;
1) a step of displaying the perspective image created in step (10).

In this configuration, steps (6) to (9)
In step (3), a three-dimensional image created based on a series of slice images is cut into a plurality of layers, and image data is projected onto a projection plane for each of the cut sections.
For the projected image data, an accumulation operation is performed for each pixel of the perspective image. For this reason, the calculation of the projection enlargement of each pixel of the section image and the calculation of the correspondence between the coordinates of each pixel of the section and the coordinates of the pixel of the perspective image can be performed for each section, and the accumulation of projection data can be performed. The calculation can be performed by accumulating the projection data for the number of cut sections on each pixel of the perspective image, so that the calculation procedure can be simplified and the calculation time can be reduced.

[0011]

Embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 shows an outline of a method of creating a fluoroscopic image using the X-ray CT apparatus of the present invention. Figure
In FIG. 1, the method for creating a perspective image of the present invention is roughly divided into five procedures shown in FIGS. 1 (a) to 1 (d). Hereinafter, these will be sequentially described.

FIG. 1A shows a first procedure, in which an X-ray C
This is a procedure in which a scan (scan) 11 of the subject 10 is performed using the T apparatus to acquire X-ray attenuation data for creating a slice image. In this procedure, the target site of the subject 10
The scan 12 is performed along the body axis direction (y-axis direction) 13 to collect X-ray attenuation data (hereinafter referred to as X-ray measurement data) for a plurality of slices.

FIG. 1B shows a second procedure, in which each slice image (hereinafter, referred to as a slice image) 20 is created by using X-ray measurement data for a plurality of slices collected in the first procedure. Procedure. In this procedure, after calculating the reconstructed image 20 of each slice, each slice image data (image data on the Xz plane) is created and stored.

FIG. 1C shows a third procedure. In this procedure, slice image data for each slice obtained in the second procedure is first stacked along the body axis direction (y-axis direction) 13. Then, image data (hereinafter, referred to as three-dimensional image data) 24 of the three-dimensional image 22 of the attention site 12 of the subject 10 is created. At this time, the image data between slices and the image data of the peripheral part of the subject are obtained by interpolation as necessary.

FIG. 1C also shows a fourth procedure. This fourth procedure is a procedure for setting a viewpoint P and a projection plane 28 as a reference for creating a fluoroscopic image of the attention site 12 of the subject 10. The viewpoint P corresponds to an X-ray source during fluoroscopy, and the projection plane 28 is a plane on which an X-ray film or an X-ray image receiving system is arranged. As the projection plane 28, a plane parallel to the X-y plane or the like set in the three-dimensional image data 24 is usually selected to facilitate the calculation.

FIG. 1 (d) shows a fifth procedure, in which the three-dimensional image data 24 of the target part 12 of the subject 10 is projected on a projection plane 28 based on a viewpoint P, and a perspective image 30 is shown. This is the procedure for creating the image data. In this procedure, 3D image data 2
4 is cut into a plurality of cut sections 31 (31A, 31B, 31C) parallel to the projection plane 26.
), And the image data 24 of each section 31 is projected onto the projection plane 28.
(The image is enlarged and projected), and the projected image data is accumulated, and the accumulated value is used as the image data of the perspective image 30.

What has been described above is an outline of the procedure of a method for creating a fluoroscopic image using the X-ray CT apparatus of the present invention.
The details will be described later according to the flowchart of FIG.

Next, a specific example in which the three-dimensional image data 24 of the target part 12 of the subject 10 is projected on the projection plane 28 will be described. FIG. 2 shows the 3D image data 24 on the projection surface 28.
1 shows an example of projection. In this projection example, the viewpoint P is a three-dimensional image 22
At the top of the image center. The three-dimensional image 22 (corresponding to the region where the three-dimensional image data 24 exists) of the attention site 12 of the subject 10 is
Each side is composed of a rectangular parallelepiped with a, b, and c. The X axis direction is a, the y axis direction (body axis direction) is b, and the z axis direction is c. Viewpoint P is the top surface of the rectangular parallelepiped (top surface parallel to the X-y plane) A
It is located at a distance h ′ from the BCD, and a perpendicular from the viewpoint P to the upper surface of the rectangular parallelepiped passes through the center Q of the upper surface of the rectangular parallelepiped. Projection surface 2
8 is arranged in parallel with the X-y plane, and the distance between the viewpoint P and the projection plane 28 is h.

It is assumed that the three-dimensional image 22 is divided into l in the X-axis direction, m in the y-axis direction, and n in the z-axis direction. Therefore, the three-dimensional image 22 includes l × m × n pixels and image data 24. The division numbers l, m, and n are usually not equal, but may be equal for easy calculation. Also, for convenience in calculation, l, m, and n are 64, 128, 256,
In many cases, a number such as 512 is used. Number of divisions in the z-axis direction n
Corresponds to the number of the sections 31 in FIG. 1 (e), and the three-dimensional image 22 has the n-layer sections 31- in the z-axis direction, that is, the projection direction.
, 31-k,..., 31-n, and the image data of the n-layer section 31 is projected onto the projection plane. Also,
Each section 31 holds image data of l × m pixels.

Next, each section of the three-dimensional image 22 will be described with reference to FIG.
The projection of the image data 24 of the image 31 (hereinafter referred to as a section image) 32 onto the projection plane 28 will be described. FIG. 3 is a view of the first projection example of FIG. 2 as viewed from the y-axis direction. In FIG. 3, with respect to the three-dimensional image 22, a plane parallel to the Xz plane can be seen. The three-dimensional image 22 is an n-layer section 31 parallel to the X-y plane.
Although they are divided into -1,..., 31-k,..., 31-n, FIG. 3 shows only components parallel to the X axis. In FIG. 3, when each section 31 is projected on the projection plane with reference to the viewpoint P, each section image 32 is projected at a different magnification. The pixel on the X-axis in the section 31-k which is the k-th layer from the top is the projection magnification αX of the following equation (1).
At k, it is enlarged and projected on the projection plane 28 as pixels of the projected image 40-k. αXk = h / (h ′ + c · k / n) (1) where k = 1, 2,..., n

Further, since the y-axis direction is also parallel to the projection plane 28, the pixels parallel to the y-axis in the cut plane 31-k, which is the k-th layer from the top, are similarly obtained by the following equation (2). ) Is enlarged and projected at the projection magnification rate αyk. αyk = h / (h ′ + c · k / n) (2) where k = 1, 2,..., n

Similarly, the sectional images 32 of the other layers are projected onto the projection plane.
The image is enlarged and projected on the projection image 40 at different magnifications. At this time, the center of each section image 32 is the projected image
It is projected to coincide with the center R of 40. That is, in the present projection example, the projection is performed so that the projection center line coincides with the perpendicular PQR from the viewpoint P.

As described above, the pixel of each section 31 is defined by the projection plane
Since the image is projected on the projection 28 at different projection magnification rates, the image is projected as a pixel having a different area for each section. For this reason,
As a result of the projection of the three-dimensional image data 24 associated with the individual pixels having the same area in the three-dimensional image 22 onto the projection plane 28, the image data 25 of the projection image 40 on which each section image 32 is projected Will be associated with pixels having different areas for each section on the projection plane 28.

For the perspective image 30 created by projecting the three-dimensional image 22 on the projection plane 28, the position, size, number of pixels, pixel size, pixel array It is necessary to set such as. The center position of the perspective image can be determined from the position of the projection plane 28, the projection center line from the three-dimensional image 22, and the like. The size of the image plane is determined by the number of pixels and the size of the pixels.
Regarding the items, if any two items are determined, the remaining items are determined, so only the two items need to be considered.

Although the number of pixels of the perspective image 30 on the projection plane 28 can be set to any number, the number of pixels is usually the same as the number of pixels of the section 31 parallel to the projection plane 28 of the three-dimensional image 22. l × m
Is set to Also in this projection example, the number is set to l × m. As described above, it is convenient for calculation if the number of pixels is the same as the number of pixels in the section. In addition, the size of the pixels of the perspective image 30 can be set to an arbitrary size, but it is appropriate to set the size of the lowermost or uppermost pixel of the three-dimensional image 22 to the size projected onto the projection plane 28. is there. In this projection example, the size of the lowermost pixel is set to the size projected on the projection plane 28. Also, a perspective image
It is computationally convenient to make the arrangement of the 30 pixels the same as that of the projected image 40. Such setting of the specifications of the perspective image 30 can be performed at the time of setting the projection plane 28 or before projecting the image data of the projection image.

When the number of pixels and the size of the pixel of the perspective image 30 are set as described above, the projection of the pixel of the section 31 of the three-dimensional image 22 onto the projection surface 28 causes the projected pixel of each section 31 to be projected. Has an area equal to or larger than the size of the pixel of the perspective image 30.

FIG. 4 is a diagram showing an array of image data on the projection plane in the first projection example. 4, a projection image 40 of a section 31 of the three-dimensional image 22 is arranged around a point R (a foot point of the perpendicular PQR) on the projection plane 28. Projected image 40-1,
…, 40-k,…, 40-n are cut sections 31-1,…, 31-k,
, 31-n, and the size of the pixel is larger in the projected image 40 corresponding to the section 31 closer to the viewpoint P.

In this projection example, the section 3 farthest from the viewpoint P
The size and the number of pixels of the projection image 40-n and the perspective image 30 corresponding to 1-n are set to be the same. Here, in forming the perspective image 30, it is necessary to sequentially accumulate the image data of the n projection images 40 with reference to the pixels of the perspective image 30.

As the image data of the projection image 40, the CT value of the CT image is used. As the reference level of the CT value,
Normally, air with -1,000 and water with 0 is used.In this projection example, when accumulating the image data of the projection image 40, the reference level of the CT value is set to 0 for air and 1,000 for water. In the case where the image data is integrated by moving to (1) and the accumulated result is a fluoroscopic image, the reference level of the CT value is returned to the original level.

The image data of the projection image 40 includes a projection plane
Each pixel (i, j) of the perspective image 30 on 28 (where (i, j)
(N is a pixel number), there are n image data corresponding to the number of layers of the section 31. For example, the image data corresponding to the pixel (i, j) on the perspective image 30 of the projection image 40-k is d2
Assuming (i, j, k), the image data D2 (i, j) of the pixel (i, j) of the perspective image 30 is expressed by Expression (3).

(Equation 1) Here, d2 (i, j, k) is the image data after shifting the reference level of the CT value, and the image data with the CT value whose reference level is restored to the original is d1 (i, j). , k).

The image data D1 (i, j) of the perspective image 30 is obtained by restoring the reference of the CT value of the image data D2 (i, j). The perspective image 30 based on the obtained image data D1 (i, j) is a perspective image (pseudo-perspective image) based on CT values.

Next, a method of obtaining the above-described image data d1 (i, j, k) or d2 (i, j, k) will be described. As described above, each pixel of the perspective image 30 and each pixel of the projection image 40-k do not usually match. Also, the size of the pixel is larger in the projection image 40 than in the perspective image 30. Perspective image 30
When there is such a relationship between the pixel of the projection image 40 and the pixel of the projection image 40, the projection image 40
The method of covering the pixels of the perspective image 30 with the pixels of FIG.
As shown in the figure, the case of covering with one pixel (FIG. 5 (a)), the case of covering with two pixels (FIG. 5 (b)) and the case of covering with four pixels (FIG. 5 (c))
There is.

In consideration of this, the image data d1 (i,
j, k) or d2 (i, j, k) are obtained as follows: (1) The image of each pixel of the projection image 40 is determined according to the ratio of the area where the pixels of the projection image 40 cover the pixels of the perspective image 30. Cumulatively add data. For example, in the case of covering with two pixels, if the area ratio of each pixel is β1, β2 and the image data of each pixel is S1, S2, the image data d to be obtained is (β1S1 + β
2S2) / (β1 + β2). (2) For each pixel of the projected image 40, if the coordinates of the center or corner of the pixel are included in the pixels of the perspective image 30,
It is assumed that the image data S of the pixel is obtained as image data d.
If there is no corresponding coordinate, the image data of the adjacent pixel is used. (3) Among the pixels of the projection image 40, the image data d for obtaining the image data S of the pixel having the largest area covering the pixel of the perspective image 30 is obtained. In the case of the same area, an average value or one of the values may be taken.

Of the above three methods, method (1) has a high calculation accuracy, but requires a long calculation time. The method (2) or the method (3) is simple and has a short operation time. When the pixels of the perspective image 30 are small, the method (2) or the method (3) is suitable. Further, in the present projection example, the description has been made assuming that the pixels of the perspective image 30 are smaller than the pixels of the projection image 40; In this case, the method of covering pixels and the method of obtaining projection image data are slightly different in detail, but the same method can be applied.

Next, a method of creating a fluoroscopic image using the X-ray CT apparatus of the present invention will be described with reference to FIGS. 6, 7 and 1. FIG. 6 is a block diagram of an X-ray CT apparatus used to create a perspective image according to the present invention, and FIG. 7 is an example of a flowchart of a method of creating a perspective image using the X-ray CT apparatus of the present invention.

First, the configuration of the X-ray CT apparatus shown in FIG. 6 will be described with reference to FIG. In FIG. 6, an X-ray CT apparatus according to the present invention includes a scanning device 61 and an image reconstruction device.
62, image storage device 63, interpolation operation device 64, position input device 6
5.Enlargement operation device 66, accumulation operation device 67, image display device 68
Is provided.

The scanning device 61 is used to scan the target region 12 of the subject 10.
Is scanned in the body axis direction 13 to obtain Xs of a plurality of slice planes.
Collect line attenuation data. The image reconstruction device 62 performs a series of slice images 20 based on the X-ray attenuation data of the slice plane.
Reconfigure. The image storage device 63 stores the region of interest 12 of the subject 10.
Is stored. The interpolation calculation device 64 performs data interpolation based on a series of slice plane image data, and creates image data 24 of the three-dimensional image 22 of the target portion 12. The position input device 65 includes coordinates of a viewpoint P for projecting the image data 24 of the three-dimensional image 22 and a projection surface 28.
The position data and the data of the structural specification of the perspective image 30 are input, and a three-dimensional image 22 section 31 for projecting the image data 24 of the three-dimensional image 22 is set. The enlargement operation device 66 displays the image of the section 31 of the three-dimensional image 22 on the projection plane.
The projection magnification rate of each section 31 when projecting the image on the projection surface 28 is calculated by equations (1) and (2). The accumulator 67
Image data of the image of each section 31 of the three-dimensional image 22 is projected onto the projection plane.
The image data of the perspective image 30 is calculated by integrating the image data projected on the projection surface 28 and the image data projected on the projection plane 28. The image display device 68 is configured to generate a perspective image based on the image data of the perspective image 30.
Display 30 on the display screen.

Next, a method of creating a perspective image using the X-ray CT apparatus of the present invention will be described based on the flowchart of the method of creating a perspective image in FIG. 7 and the outline of the method of creating a perspective image in FIG. The flowchart of FIG. 7 corresponds to the X-ray CT of FIG.
It is performed using the device. In the first step (101),
Using a scanning device 61 of the X-ray CT apparatus, a scan 11 of a plurality of slice planes is performed on a region of interest 12 of the subject 10 as shown in FIG. 1A, and scan data of a plurality of slice planes is collected. For example, when the abdomen or chest of the subject 10 is set as the region of interest 12, usually about 30 to 50 scans, a maximum of 100 scans
The X-ray attenuation data of the scan is collected.

In the second step (102), based on the series of scan data collected above, the image reconstruction device 62
, The subject 10 is scanned every scan as shown in FIG.
The reconstruction operation of the slice image 20 of the slice plane is performed. This second step (102) is usually performed simultaneously with the first step (101). If scan data is stored in advance, the process may be started from this step.

In the third step (103), the image data of the slice image 20 calculated in the second step (102) is collected for all of the series of scans of the region of interest 12 and stored in the image storage device 63. This third step (103) may be performed together with the second step (102).

In a fourth step (104), based on the image data of a series of slice images 20 collected in the third step (103), necessary image data between the slice images 20 is Obtained by interpolation calculation. By this data interpolation calculation, image data 24 of a three-dimensional image 22 of the target part 12 of the subject 10 as shown in FIG.

The three-dimensional image data 24 is usually rectangular parallelepiped data, and the coordinate system thereof is such that the body axis direction is the y-axis and the Xz plane is perpendicular to the body axis direction. Therefore, the slice plane is parallel to the Xz plane, and the slice image
20 image data are obtained as image data of a plane parallel to the Xz plane. Therefore, in the fourth step (104),
Based on the image data of the series of slice images 20, data interpolation is performed mainly in the y-axis direction. Less than,
The reason why the data interpolation calculation is performed in the y-axis direction and the pixel size in each coordinate system will be described.

The density of the rectangular parallelepiped three-dimensional image data 24 or the size of the pixel changes depending on the size of the effective field of view (hereinafter abbreviated as FOV) at the time of reconstructing the slice image in the second step (102). In the X-axis and z-axis directions, since the image data of the slice image 20 is often used as it is, the pixel size of the three-dimensional image data 24 in that direction is the same as the pixel size of the slice image 20. The size of one pixel (mm) = FOV (mm) / 512 (the number of pixels). The number of pixels is not limited to 512, but 64, 128, 25
6, 1024 and so on.

On the other hand, the y-axis direction varies depending on the position of the slice plane (corresponding to the position of the slice image), that is, the interval at which the slice images are arranged. Therefore, the size of a pixel in the y-axis direction is usually one pixel size (mm) =
The measurement range (corresponding to the range of the attention site 12) (mm) / the number of reconstructed images = reconstructed image pitch (mm).

As described in the first step (101), the number of reconstructed images is usually 30 to 50 (up to 100).
In the axial direction, the number of pixels is usually smaller than in the other axial directions. Therefore, the pixel size in the y-axis direction is also
It is larger than in the axial direction. Therefore, the image data between the slice images 20 is obtained by the interpolation calculation in this step (104), and the pixel size and the number of pixels are equivalent to those in the X-axis and z-axis directions. Create 24.

In this step (104), missing image data of the image data in the X-axis direction and the z-axis direction is also obtained by interpolation and compensated to complete the rectangular parallelepiped three-dimensional image data 24.

In the fifth step (105), as shown in FIG. 1 (d), the viewpoint P and the projection plane 28 for setting the perspective image 40 from the image data 24 of the three-dimensional image 22 (hereinafter, these settings are referred to as Initial value setting). This initial value setting is
Using the position input device 65, the coordinates of the viewpoint P and the data of the position of the projection plane 28 are input. Hereinafter, description will be made with reference to a first projection example shown in FIG.

In FIG. 2, the three-dimensional image 22 of the rectangular parallelepiped is X
-It is arranged with the plane parallel to the y-plane as the upper surface and the z-axis as the vertical direction. The viewpoint P is on the vertical line of the center Q of the top surface ABCD of the three-dimensional image 22, and is located at a distance h 'upward. The projection plane 28 is arranged at a position away from the viewpoint P by a distance h and perpendicular to the z-axis. Regarding the setting of the viewpoint P, the distance h 'is usually set to a distance equivalent to the FOV of the three-dimensional image 22 so that the three-dimensional image 22 is appropriately projected. Therefore, in the present projection example, the distance h ′ has a dimension close to a or b. However, the distance h 'is not limited to this, and can take various dimensions in consideration of the projection magnification of the three-dimensional image 22. The position of the projection plane 28 is usually the lowermost surface (the surface farthest from the viewpoint P) EFGH of the three-dimensional image 22.
Or a position slightly away from it.

In the sixth step (106), a section 31 is set in the z-axis direction of the three-dimensional image 22 of a rectangular parallelepiped. The setting of the section 31 is also performed at the position input position 65. Cross section 31
Is normally set to 512. This is in accordance with the number of pixels in the z-axis direction, since the image data of the slice image 20 is normally displayed in 512 × 512 pixels. Further, the fact that the section 31 is parallel to the X-y plane of the three-dimensional image 22 of the rectangular parallelepiped means that even when the viewpoint P and the projection plane 28 move or rotate, the rectangular parallelepiped (three-dimensional image) Each of the 22 axes is unchanged.

In the seventh step (107), the three-dimensional image 22
Image 32 (hereinafter referred to as a section image) of the section 31 of the projection plane
The projection enlargement ratio for projecting onto 28 is calculated. The calculation of the projection enlargement ratio of the section image 32 is performed by the enlargement operation device 66. The projection magnification rate of the cross-section image 32 changes depending on the position of the cross-section 31 and is calculated in both directions of the X-axis and the y-axis by the equations (1) and (2). At this time, each section
Reference numeral 31 denotes a rectangular parallelepiped (three-dimensional image) 22 which remains parallel to the X-y plane, but the size of the pixel projected on the projection surface 28 is
It differs every 31. This means that the projection magnification of the section image 32 is
This is because it differs for every 31. It is a point to be careful in integrating image data.

In the eighth step (108), the specifications of the perspective image created on the projection plane 28 are set. This perspective image 30
May be performed at the time of setting the projection plane 28 in the fifth step. The specification of the perspective image 30 is a specification of how to create the perspective image 30 on the projection plane 28, and the coordinates of the center position of the perspective image 30 (not limited thereto, and defines the arrangement of the perspective image 30). Other coordinates may be used as the reference coordinates), the number of pixels, the size of the pixels, the arrangement of the pixels, and the like. In setting the specifications of the perspective image 30, data is input using the position input device 65.

In this embodiment, the coordinates of the center position of the perspective image 30 are set to the point R in FIG. The number of pixels and the arrangement of the pixels are the same as the lowermost surface 31-n of the section 31, and the pixel size is the same as that of the pixel when the pixel of the lowermost surface 31-n of the section 31 is projected onto the projection surface 28. It is the same as the size. This is the perspective image
By setting the specifications of 30, the size of the obtained perspective image 30 is the size obtained by projecting the lowermost surface 31-n of the cut surface 31 onto the projection surface. The above setting example of the specifications is for facilitating the calculation, and is not limited thereto, and various specifications can be set.

In the ninth step (109), the image data 24 of the section image 32 is projected on the projection plane 28. A CT value is used as the image data 24 and is projected for each pixel. At this time, each pixel is enlarged and projected at a different projection enlargement ratio for each position of the cut plane 31, and the difference in the size of the pixel is replaced by the difference in the positional relationship of the cut plane 31 from the viewpoint P. So you can ignore it.

In projecting the image data 24 of the section image 32, a process of moving the reference level of the CT value of the image data 24 is performed. That is, for all image data 24, CT
Add the value 1,000, air level 0, water level 1,00
The reference level is moved so that it becomes zero. After the reference level shift processing of the CT value, the image data 24 of the section image 32 is projected on the projection plane 28. Of the image data after this processing,
Set to 0 for negative image data.

In the tenth step (110), the image data 25 of the projection image 40 projected on the projection surface 28 is sequentially accumulated.
This is performed by using the accumulative calculation device 67 of the image data 25 of the projection image 40. In the cumulative calculation of the image data 25 of the projection image 40, since the size of the pixel of the projection image 40 and the size of the pixel of the perspective image 30 are generally different, it is necessary to align the pixels of the two images as described above (see FIG. 5). Here, in the case of (2), for each pixel of the projected image 40, if the coordinates of the center of the pixel are included in the pixel of the perspective image 30, the image data d2 (i, j, k ). "Method is adopted. After performing such pixel alignment, the image data d2 (i, j, k) of the projection image 40 for n layers to be included in each pixel of the perspective image 30
Is cumulatively calculated by Expression (3). 1 / n on the right side of equation (3)
Is weighting.

As a result of this accumulation operation, the perspective image 30 is calculated as a weighted accumulated value of the image data d2 (i, j, k) of the projection image 40.
Image data D2 (i, j) is obtained. The order of accumulating the image data is not particularly specified, and is performed for the image data of all the projection images 40.

The image data D2 of the fluoroscopic image 30 obtained above
Since (i, j) is obtained by moving the reference value of the CT value by 1,000, the reference value of the CT value is re-moved with respect to the image data D2 (i, j) to return to the original CT value. Need to be kept. The image data D1 (i, j) of the fluoroscopic image 30 is obtained by re-moving the CT value at the reference level.

In the eleventh step (111), the perspective image 30
A fluoroscopic image 30 is displayed based on the image data D1 (i, j). The display of the perspective image 30 is performed using the image display device 68. The image displayed here is based on the CT value.

The perspective image 30 obtained in the present embodiment has the viewpoint P
Of the X-ray attenuation as viewed from above, resulting in a fluoroscopic image like a normal radiograph. In this case, at the time of magnifying projection calculation of the cross-sectional image 31, a threshold value is provided for image data, image data outside the threshold value is excluded, or a calculation range is set on the original image data. can do.

FIGS. 8 and 9 show three-dimensional image data 24.
2 shows a second example of projection onto the projection plane 28. FIG. 8 is a view for explaining a second example of projection of three-dimensional image data onto a projection plane.
FIG. 9 is a diagram showing an array of image data on the projection plane in the second projection example. As can be seen from FIG. 8, in this projection example, the viewpoint P is located at a position away from the central axis of the three-dimensional image 22. However, the viewpoint P is located above the uppermost surface of the three-dimensional image 22, and the projection plane 28 is arranged parallel to the uppermost surface of the three-dimensional image 22. In this case, a perpendicular is dropped from the viewpoint P to the projection plane 28, and the perpendicular is drawn on the top surface of the three-dimensional image 22 and the projection plane 28.
If the points intersecting are denoted by Q and R, respectively, and projection of the image data 24 is performed with reference to the perpendicular PQR, the calculation of the perspective image 30 is simplified.

In FIG. 8, the distances from the viewpoint P to the projection plane 28 and the uppermost plane of the three-dimensional image 22 are h and h ′, respectively.
Therefore, each section 31-1,..., 31-k of the three-dimensional image 22
, 31-n are represented by Expressions (1) and (2) as in the first projection example. Regarding the size of the image plane, the number of pixels, the pixel size, and the pixel arrangement of the perspective image 30,
The section 31-n on the lowermost surface of the two-dimensional image 22 is set to be the same as the projection image 40-n projected on the projection plane 28.

Further, if the reference point of the pixel array of the perspective image 30 is set to the point R shown in FIG. 9, the computation procedure of the first projection example can be applied as it is to compute the image data of the perspective image 30. . In this case, since the coordinates of the point R are shifted from the center position as compared with FIG. 4, it is sufficient to move the coordinates (i, j) of the pixel by the shifted amount.

Next, the third projection of the three-dimensional image data onto the projection plane is performed.
Will be described with reference to FIG. In this projection example, the projection plane 28 has an angle θ with respect to the section 31 of the three-dimensional image 22.
Just leaning. FIG. 10 shows a three-dimensional image in this projection example.
FIG. 9 shows the relationship between the point 22, the viewpoint P, and the projection plane 28, and is a diagram viewed from the y-axis direction. In the case of this projection example, since the projection plane 28 is inclined by the angle θ with respect to the three-dimensional image 22, when the section plane image 32 is projected on the projection plane 28,
Varies depending on the position. FIG.
Here, this state is shown by taking the section 31-k as an example.

In FIG. 10, a section 31-k is a section parallel to the X-axis. When a perpendicular is lowered from the viewpoint P to the projection plane 28, assuming that the intersection with the projection plane 28 is R and the intersection with the section 31-k is Qk, the point Qk of the section 31-k is projected to the point R. Will be. Further, when comparing the left end pixel and the right end pixel of the cut plane 31-k, the right end pixel is projected with a larger projection magnification. As shown in FIG. 10, the distances from the plane to which the viewpoint P belongs to the projection plane 28 and the left and right ends of the cut plane 31-k are h, hl, and hr, respectively, and the length of the cut plane 31 in the X-axis direction is
a, if the number of pixels in the X-axis direction is 512, for example,
The projection enlargement ratio of each pixel above is expressed by Expression (4) or Expression (5). αXk = h / (hl−aisin θ / 512) (4) = h / {hl (512−i) / 512 + hri / 512} (5) where i is a cross section 31 The coordinates of the pixel of -k in the X-axis direction.

If the distance from the plane to which the viewpoint P belongs to the intersection Qk of the section 31-k is hk, and the coordinates of the pixel including the intersection Qk in the X-axis direction are Xk, each pixel on the section 31-k Can be expressed by equation (6). αXk = h / {hk + a (Xk−i) sin θ / 512} (6) where i is the coordinate of the pixel of the section 31-k in the X-axis direction.

The specification of the perspective image 30 in this projection example can be arbitrarily set on the projection plane 28, but the number of pixels, the size of the pixels, and the arrangement of the pixels are set based on the three-dimensional image 22. Is computationally convenient. When setting the coordinates of the pixels of the perspective image 30, it is convenient to use the point R of the foot perpendicular to the viewpoint P as a reference (for example, the origin). Since all the pixels on the perpendicular PR are projected onto a point R on the projection plane 28, it is convenient for calculation to use the intersection Qk with the perpendicular PR as the coordinate reference point for each section plane 31. In this case, it is advantageous to use the expression (6) for the calculation of the projection magnification rate.

As for the other steps, the accumulative calculation of the image data for each pixel of the perspective image 30 can be performed by following the procedure described in the first and second projection examples.

Further, as another projection example, the projection plane 28 is 3
When the image is inclined in both the X-axis direction and the y-axis direction with respect to the two-dimensional image 22, the processing for the X-axis direction in the third projection example is also applied to the y-axis direction, so that the projection is enlarged. The calculation of the rate and the cumulative calculation of the image data become possible. In other steps, image data of a perspective image can be obtained by following the procedure described in the first projection example.

[0069]

As described above, according to the present invention, X
Based on a series of slice images of a region of interest of a subject reconstructed using a line CT apparatus, a procedure for creating a fluoroscopic image of a region of interest of the subject is simplified, and the time for the creation process is reduced. You.

[Brief description of the drawings]

FIG. 1 is an outline of a method for creating a fluoroscopic image using an X-ray CT apparatus of the present invention.

FIG. 2 is a view for explaining a first example of projection of three-dimensional image data on a projection surface.

FIG. 3 is a view of FIG. 2 viewed from a y-axis direction.

FIG. 4 is a diagram showing an array of image data on a projection plane in a first projection example.

FIG. 5 shows how to cover pixels of a perspective image with pixels of a projection image.

FIG. 6 is a block diagram of an X-ray CT apparatus used to create a fluoroscopic image according to the present invention.

FIG. 7 is a flowchart of a method for creating a fluoroscopic image using the X-ray CT apparatus of the present invention.

FIG. 8 is a view for explaining a second example of projection of three-dimensional image data onto a projection surface.

FIG. 9 is a diagram showing an array of image data on a projection plane in a second projection example.

FIG. 10 is a view for explaining a third example of projection of three-dimensional image data onto a projection plane.

[Explanation of symbols]

 10 subject 11 scan (scan) 12 region of interest 13 ... body axis direction (y-axis direction) 20 ... slice reconstruction image (slice image) 22 ... three-dimensional image 24 ... three-dimensional image data 25 ... projection image Image data 28 ... projection plane 30 ... perspective image 31 ... section 32 ... section image 40 ... projection image 61 ... scanning device 62 ... image reconstruction device 63 ... image storage device 64 ... interpolation calculation device 65 ... position input device 66 … Enlargement calculation device 67… Cumulative calculation device 68… Image display device

Claims (2)

[Claims] 1. A scanner comprising an X-ray source and an X-ray detector opposed to each other, and an image reconstructing apparatus for reconstructing a plurality of slice images of the subject based on scan data of the subject collected by the scanner. In an X-ray CT apparatus, an image storage device that collectively collects and accumulates a series of slice image data, and an interpolation operation that obtains image data of a three-dimensional image by performing an interpolation operation on image data that lacks a series of slice images A position input device for inputting data for setting a viewpoint position and a projection plane for projecting a three-dimensional image and data for setting a structural specification of a perspective image created on the projection plane; A magnifying device for calculating a projection magnification of each image when projecting a cross-sectional image obtained by cutting a three-dimensional image, performing a magnifying operation of the size of each pixel and calculating a projection coordinate, and a perspective image of It is characterized by comprising an accumulator which accumulates projection data of a cross-sectional image projected thereon for each pixel to create image data of a perspective image, and an image display device which displays a perspective image. X-ray CT device. 2. An object is scanned by an X-ray CT apparatus,
In a method for creating a fluoroscopic image based on a series of slice images obtained by image reconstruction and placing a viewpoint at an arbitrary spatial position, a fluoroscopic image is generated. (2) a step of scanning a target region and collecting scan data, (2) a step of reconstructing a slice image of a plurality of slices based on the scan data collected in step (1), and (3) a step (2). (4) collect and accumulate the image data of the slice images obtained in step (1) as a series of slice image data, and (4) find missing data by interpolation based on the series of slice image data to obtain a three-dimensional (5) setting a viewpoint position and a projection plane for projecting a three-dimensional image, and (6) three-dimensional image
A step of setting a section of a plurality of layers by cutting in parallel with a plane facing the viewpoint, (7) a step of calculating a projection magnification of each pixel of the section image, and (8) a step of setting a projection plane. A step of setting the structural specifications of the perspective image to be created;
(9) Step of projecting image data on each pixel of the perspective image onto the pixel of the perspective image, and (10) Accumulation of projection data of the section image for each pixel of the perspective image to create image data of the perspective image Step (1) and Step (1)
A step of displaying the perspective image created in step 0).