JP3204886B2 - X-ray CT system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an X-ray CT (Com
More specifically, a detector array in which a large number of X-ray detectors are arranged in a line.
The present invention relates to an X-ray CT apparatus that collects data by a helical scan using a multistage detector array in which two or more stages are arranged side by side and generates an image.

[0002]

2. Description of the Related Art Conventionally, an object to be scanned is relatively linearly moved along one axis with respect to a detector array in which a number of X-ray detectors are arranged in a line, and an X-ray tube is arranged around the object to be scanned. A data collection unit that collects data while rotating (this is referred to as a helical scan), and data of a plurality of views necessary to generate an image at one position on the linear movement axis are provided in the vicinity of the image generation position. 2. Description of the Related Art An X-ray CT apparatus is known that includes a data calculation unit that calculates by interpolation from collected data and an image generation unit that generates an image from the data of the plurality of views calculated by the data calculation unit. The data calculation means selects two data, which are data of the same view as each of the plurality of views or data of the opposite view, and are closest to the image generation position from the collected data, perform an interpolation operation, and calculate an image. Calculate data to generate.

[0003] Conventionally, the detector array is set to N (≧
2) A multi-stage detector array is provided side by side, and data of a plurality of views necessary for generating an image while rotating an X-ray tube around a scan target is collected by each detector array (this is a normal scan). An X-ray CT apparatus is known which includes data collection means and image generation means for respectively generating a plurality of images corresponding to each detector array from data of a plurality of views collected by each detector array. .

[0004]

As described above, there is known an X-ray CT apparatus which collects data by helical scan using a single-stage detector array and generates an image.
2. Description of the Related Art An X-ray CT apparatus that collects data by a normal scan using a multistage detector array and generates an image is known. However, there is no known X-ray CT apparatus that collects data by helical scan using a multistage detector array and generates an image. Therefore, an object of the present invention is to provide an X-ray CT apparatus that collects data by helical scan using a multistage detector array and generates an image.

[0005]

According to a first aspect of the present invention, there is provided a scanning target for a multi-stage detector array in which a plurality of X-ray detectors arranged in a row and two or more stages are arranged in parallel. Data acquisition means for relatively moving the X-ray tube relatively along the axis in the juxtaposition direction and collecting data while rotating the X-ray tube around the object to be scanned, and an image at one position on the linear movement axis Data calculation means for calculating data of a plurality of views necessary for generating the data by interpolation from data collected in the vicinity of the image generation position, and generating an image from the data of the plurality of views calculated by the data calculation means An X-ray CT apparatus including an image generation unit, wherein the data calculation unit is the same view data or the opposite view data as each of the plurality of views and the image generation position. Interpolation calculating the two closest data by selecting from the data collected at each detector array, to provide an X-ray CT apparatus characterized by calculating the data for generating an image.

In the X-ray CT apparatus according to the first aspect,
Data of the same view or data of the opposite view as each of the multiple views required to generate the image,
Two data closest to the image generation position are selected from the data collected by each detector array, and an interpolation operation is performed using the two data to calculate data for generating an image. That is, at the time of interpolation calculation, two data closest to the image generation position are selected and used, even if the data is collected by different detector arrays. For this reason, it is possible to use two data that are closer to the image generation position than when the two data closest to the image generation position are selected and used from among the data collected by the same detector array. Can be improved.

[0007] In a second aspect, the present invention relates to a multi-stage detector array in which a large number of X-ray detectors are arranged in a row and two or more X-ray detectors are arranged in parallel. Data collection means for relatively linearly moving along an axis and collecting data while rotating an X-ray tube around a scan target; and necessary for generating an image at one position on the linear movement axis Data calculation means for calculating data of a plurality of views from data collected near the image generation position by interpolation, and image generation means for generating an image from the data of the plurality of views calculated by the data calculation means. An X-ray CT apparatus, wherein the data calculating means includes two data which are the same view data or the opposite view data of each of the plurality of views and which are closest to the image generation position. Data is selected from the data collected by each detector array, and the thickness of the fan-shaped X-ray beam incident on each detector array is set to th, and the thickness is set relative to each rotation of the X-ray tube or multistage detector array. When the distance of linear movement is d and p = d / th is a helical pitch, an interpolation coefficient is determined based on the helical pitch p, an interpolation operation is performed using the interpolation coefficient and the selected data, and an image is calculated. An X-ray CT apparatus is provided, which calculates data for generating an image.

In the X-ray CT apparatus according to the second aspect,
Data of the same view or data of the opposite view as each of the multiple views required to generate the image,
The two data closest to the image generation position are selected from the data collected by each detector array. Further, an interpolation coefficient is determined based on the helical pitch p. Then, an interpolation operation is performed using the interpolation coefficient and the two selected data to calculate data for generating an image. That is, at the time of interpolation calculation, two data closest to the image generation position are selected and used, even if the data is collected by different detector arrays. Further, an interpolation coefficient is determined based on the helical pitch p. For this reason, it is possible to use two data that are closer to the image generation position than when the two data closest to the image generation position are selected and used from among the data collected by the same detector array. Can be improved.

In a third aspect, the present invention relates to a multi-stage detector array in which a large number of X-ray detectors are arranged in a row and two or more stages are arranged in parallel. Data collection means for relatively linearly moving along an axis and collecting data while rotating an X-ray tube around a scan target; and necessary for generating an image at one position on the linear movement axis Data calculation means for calculating data of a plurality of views from data collected near the image generation position by interpolation, and image generation means for generating an image from the data of the plurality of views calculated by the data calculation means. The thickness of the fan-shaped X-ray beam incident on each detector array is defined as th, and the distance that the X-ray tube or the multistage detector array relatively linearly moves for each rotation is defined as d. When p = d / th is a helical pitch, the data collection unit allows the operator to select any one of a plurality of preset ps, and collects data at the selected helical pitch p; The data calculating means calculates and stores an interpolation coefficient for each of a plurality of preset ps in advance, and stores data of the same view or data of an opposite view as each of the plurality of views and the image generation position. X is selected from among the data collected by each detector array from the data collected by each detector array, and an interpolation operation is performed using the stored interpolation coefficients to calculate data for generating an image. A line CT apparatus is provided.

In the X-ray CT apparatus according to the third aspect,
Data of the same view or data of the opposite view as each of the multiple views required to generate the image,
Two data closest to the image generation position with the image generation position interposed therebetween are selected from the data collected by each detector array, and interpolation calculation is performed using the two data to generate data for generating an image. calculate. That is, at the time of interpolation calculation, two data closest to the image generation position are selected and used, even if the data is collected by different detector arrays. For this reason, it is possible to use two data that are closer to the image generation position than to select and use the two data that are closest to the image generation position from among the data collected by the same detector array. Can be improved. Further, a plurality of preset helical pitches p (for example, p = 1.5, p = 2 and p = 3)
Is calculated and stored in advance, and the operator is allowed to select one of the plurality of helical pitches p. In the actual interpolation calculation, it is sufficient to read out and use the stored interpolation coefficient. Since it is not necessary to calculate the interpolation coefficient each time, the processing time can be reduced.

In another aspect, the present invention relates to a multi-stage detector array in which a large number of X-ray detectors are arranged in a row and two or more X-ray detectors are arranged in parallel. The plurality of views necessary to collect data while moving the X-ray tube relatively linearly along the axis of the scan and rotating the X-ray tube around the scan target, and to generate an image at one position on the linear movement axis Of the same view or data of the opposite view, and two data closest to the image generation position are selected from data collected by each detector array, and a fan-shaped X-ray beam incident on each detector array is selected. , The distance linearly moving relative to each rotation of the X-ray tube or the multi-stage detector array is d, and p = d /
When th is a helical pitch, an interpolation coefficient is determined based on the helical pitch p, and interpolation calculation is performed using the interpolation coefficient and the selected data to calculate data of a plurality of views necessary for generating an image. The present invention also provides an image generation method characterized by generating an image from the calculated data of a plurality of views.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in more detail based on embodiments of the present invention shown in the drawings. In addition,
This does not limit the present invention.

First Embodiment FIG. 1 is a block diagram of an X-ray CT apparatus 100 according to a first embodiment of the present invention. This X-ray CT apparatus 100
Has an operation console 1, an imaging table 10, and a scanning gantry 20. The operation console 1
Is an input device 2 for receiving operator's instructions, information, etc.
, A central processing unit 3 for executing helical scan processing, image reconstruction processing, and the like;
0 and control interface 4 to output to scanning gantry 20
A data collection buffer 5 for collecting data acquired by the scanning gantry 20, a CRT 6 for displaying images and the like,
It has a storage device 7 for storing various data and programs. The imaging table 10 moves the subject on the body axis direction. The scanning gantry 20 includes an X-ray tube 30, a collimator 50, and a two-stage detector array 60.
An X-ray controller 21 for adjusting the timing and intensity of X-ray irradiation, a collimator controller 22 for adjusting the width and position of the X-ray transmission slit of the collimator 50,
A data collection unit 23 and an X-ray tube 30 around the body axis of the subject
And a rotation controller 24 for rotating the two-stage detector array 60 and the like.

As shown in FIG. 2, a two-stage detector array 60
Is a combination of a first-stage detector array 61 and a second-stage detector array 62. The detector array 61 of the first stage is obtained by arranging X-ray detectors 61 (γ) of many channels in an arc shape. Similarly, the detector array 62 of the second stage is obtained by arranging the X-ray detectors 62 (γ) of many channels in an arc shape. Here, γ is a channel angle, and the X-ray tube 30 and the X-ray detector
1 (γ) is an angle formed by a straight line (L in FIG. 3) connecting the X-ray tube 30 and the center of the two-stage detector array 60.

As shown in FIG. 3, the X-ray emitted from the X-ray tube 30 is converted into a flat fan-shaped X-ray beam Xr by the collimator 50, and the first-stage detector array 61 of the two-stage detector array 60 is used. And the second stage detector array 62. A straight line L connecting the X-ray tube 30 and the center of the two-stage detector array 60 is called an angle reference axis. In the X-ray detectors 61 (0) and 62 (0) at the center of the two-stage detector array 60, the channel angle γ = 0, and the X-ray at the right end in the drawing of the two-stage detector array 60 is shown.
The channel angle γ = + γm in the line detectors 61 (+ γm) and 62 (+ γm), and the channel angle in the X-ray detectors 61 (−γm) and 62 (−γm) at the left end of the two-stage detector array 60 in the drawing. γ = −γm.

FIG. 4 is an explanatory diagram of the view angle. The angle θ formed by the angle reference axis L and the vertical axis at one angular position where the X-ray tube 30 and the two-stage detector array 60 are rotated is referred to as an absolute view angle. When the absolute view angle θ at the image generation position is θ = θo, βi = θ−θo is referred to as a relative view angle. As shown in FIG. 4A, data collected at the relative view angle βi is represented by D (βi, γ).
At this time, adjacent data in the same view as the data D (βi, γ) are D (βi−2π, γ) and D (βi + 2).
π, γ). The adjacent data in the opposite view of the data D (βi, γ) is D (β-π-2γ, -γ) and D (βi + π-2γ, -γ). FIG. 4 (b)
Are adjacent data D (βi−π + 2γm, + γm) and D (βi + π +) of the opposite view of D (βi, −γm).
2γm, + γm).

FIG. 5 shows the relative view angle β and the linear movement axis z.
It is explanatory drawing of an upper position. The linear movement axis in the helical scan is the z-axis, and as shown in FIG. 5B, the intermediate position between the detector arrays 61 and 62 at a certain absolute view angle θs is the image generation position, and this is z = 0. . The thickness of the fan-shaped X-ray beam Xr incident on each of the detector arrays 61 and 62 is represented by th, and the distance that the X-ray tube 30 and the two-stage detector array 60 move linearly for each rotation is represented by d.
When p = d / th is the helical pitch, the helical pitch p = 1 in FIG.

As shown in FIG. 5A, at a relative view angle βi = −π / 2, data D1 (−π / 2, γ) obtained by the first detector array 61 has a value of z = 0. The data D2 (−π / 2, γ) obtained by the second detector array 62 is the data on the scan plane at z = −th. Also, as shown in FIG.
At the relative view angle βi = 0, the data D1 (0, γ) obtained by the first detector array 61 is the data of the scan plane of z = th / 2, and the data obtained by the second detector array 62 D2 (0, γ) is data on the scan plane at z = −th / 2. Further, as shown in FIG. 5C, at the relative view angle βi = π / 2, the data D1 (π / 2, γ) obtained by the first detector array 61 is z
= Th scan plane data, and the data D2 (π / 2, γ) obtained by the second detector array 62 is z = 0 scan plane data. Further, as shown in FIG. 5D, at a relative view angle βi = 2π, the data D1 (2π, γ) obtained by the first detector array 61 is z
= 3 · th / 2 scan plane data, and the data D2 (2π, γ) obtained by the second detector array 62 is z
= Th / 2 scan plane data. The data D1 (0, γ) and D2 (2π, γ) have the same z = th
/ 2 scan plane data and relative view angle β
Since i = 0 and βi = 2π are the same as the view β of the data for generating the image, D1 (0, γ) = D2
(2π, γ).

In general, there is an N-stage detector array, and the relative view angle bn when the data of the scan plane of z = 0 can be obtained by the n-th (= 1,..., N) detector array is as follows. bn = (2n−N−1) π / p.

FIG. 6 is a flowchart of the helical scan process. In step S1, a free mode in which the operator can freely set the helical pitch p, or p =
1.5, p = 2 or p = 3 is determined to be the fixed mode for the operator to select. If the operator specifies the free mode, the process proceeds to step S2, where the operator sets the fixed mode. If so, the process proceeds to step S3. The free mode has an advantage that an arbitrary helical pitch p can be set, but has a disadvantage that it takes a long processing time to calculate the weight of the interpolation operation and to select data used for the interpolation operation. Conversely, the fixed mode has a disadvantage that an arbitrary helical pitch p cannot be set, but the weight of the interpolation calculation is calculated and stored in advance,
Since the method of selecting the data used for the interpolation calculation is known in advance, there is an advantage that the processing time is short.

In step S2, each detector array 61,
The thickness th of the fan-shaped X-ray beam Xr incident on the beam 62, the distance d linearly moving for each rotation, and the helical pitch p
The operator sets any two of the three parameters. If two parameters are set, the other is p
= D / th. Then, the process proceeds to step S5.

In step S3, a fine pitch (p = 1.
5) The operator selects one of standard pitch (p = 2) and high-speed pitch (p = 3). In step S4, the operator sets one of the two parameters, th and d. The other one is calculated by p = d / th. Then, the process proceeds to step S5.

In step S5, the operator is caused to set scan parameters other than th, d, and p (for example, a scan start position and a scan end position). Step S6-
In S9, according to the set parameters, the imaging table 10 is moved linearly along the z-axis, and the X-ray tube 30 is moved.
While rotating the two-stage detector array 60, data is collected from the scan start position to the scan end position.

FIG. 7 is a flowchart of the image reconstruction processing. In step R1, the helical pitch p = 1.
It is determined whether data has been collected in any of 5, 2, and 3. If data is collected at any of p = 1.5, 2, and 3, the process proceeds to step R2; otherwise, the process proceeds to step R2.
Proceed to 3. In step R2, an interpolation coefficient set used for the interpolation calculation is calculated. The calculation process of the interpolation coefficient set will be described later with reference to FIG. In step R3, the data of the helical pitch for which data has been collected is read out from the interpolation coefficient sets of p = 1.5, 2, and 3 which have been calculated and stored in advance. The process of reading the interpolation coefficient set will be described later with reference to FIG. In step R4, data of a plurality of views necessary for generating an image is calculated using the collected data and the interpolation coefficient set. The data calculation processing by this interpolation calculation will be described later with reference to FIGS. In step R5, an image is generated by a known image reconstruction operation.

FIG. 8 shows an interpolation coefficient set calculation process (FIG. 7).
It is a flowchart of step R2). Step C
In step 1, it is determined whether or not p ≦ 1.
Otherwise, go to step C3. In Step C2, in the range of -π / p ≦ βi <−π / p + 2π, the interpolation coefficient w1 (βi, γ) = (π / p−βi) / (2π / p) (1) γ) is calculated. And
Go to step R4 in FIG. The basis of equation (1) is shown in FIG.
This will be described next with reference to FIG. FIG. 12 shows that p = 0.
5, FIG. 13 shows a case where p = 1, but if these are generalized, the above-mentioned step C2 is performed.

FIG. 12 is an explanatory diagram of a method of calculating an interpolation coefficient when p = 0.5. FIG. 12A shows that β = 0,
The range of the slice thickness corresponding to the data D1 (β) and D2 (β) at 2π and 4π is shown. (B) of FIG.
Indicates the position on the z-axis of the scan plane of the data D1 (β) and D2 (β) by a cosine curve. Data D1
The relative view angle of (β) is β (D1), and the data D2
The relative view angle of (β) is β (D2). Data D
Even if attention is paid only to one of 1 (β) and D2 (β), there are data for two rotations (4π) near the position z = 0 where the image is generated. Therefore, the data D1 (β), D
Using only one of 2 (β), a known full recon
It can be understood that an interpolation operation (hereinafter, referred to as two-rotation linear interpolation) for obtaining data of a plurality of views required for (construction) may be performed.

FIG. 12C shows data D 1 (β), D
The position on the z-axis of the scan plane 2 (β) is indicated by a straight line. Focusing on the data D1 (β), -4π ≦ β <
The range A of 0 is data of two rotations (4π minutes) near the position z = 0 where the image is generated, and the data D1 of the range A
It can be seen that known two-rotation linear interpolation may be performed using (β). That is, for the data D1 (β) in the range A, an interpolation coefficient may be calculated by a known interpolation coefficient calculation method of two-rotation linear interpolation. Equation (1) is not excluded from the method of calculating the interpolation coefficient of the known two-rotation linear interpolation. Similarly,
Focusing on the data D2 (β), the range B of 0 ≦ β <4π
Is equivalent to two rotations (4π) near the position z = 0 where the image is generated.
), And it can be seen that the known two-rotation linear interpolation may be performed using the data D2 (β) in the range B. That is, for the data D2 (β) in the range B,
The interpolation coefficient may be calculated by a method of calculating the interpolation coefficient of the rotation linear interpolation.

FIG. 12D shows data D1 in the range A.
FIG. 9 is an explanatory diagram of a correction coefficient for (β). -2π ≦ β
The correction coefficient for data D1 (β) in the range of <0 is w
When D1, data D1 in the range of −4π ≦ β <−2π
Since the correction coefficient for (β) can be easily obtained by (1-w1), the data D in the range of −2π ≦ β <0 is actually obtained.
Only the correction coefficient w1 for 1 (β) needs to be calculated.

FIG. 13 is an explanatory diagram of a method of calculating an interpolation coefficient when p = 1. FIG. 13A shows that β = 0,2
The range of the slice thickness corresponding to the data D1 (β) and D2 (β) at π and 4π is shown. (B) of FIG.
Indicates the position on the z-axis of the scan plane of the data D1 (β) and D2 (β) by a cosine curve. Data D1
The relative view angle of (β) is β (D1), and the data D2
The relative view angle of (β) is β (D2). Data D
Even if attention is paid only to one of 1 (β) and D2 (β), there are data for two rotations (4π) near the position z = 0 where the image is generated. Therefore, the data D1 (β), D
It can be seen that it is sufficient to perform known two-rotation linear interpolation using only one of 2 (β).

FIG. 13C shows data D1 (β), D
The position on the z-axis of the scan plane 2 (β) is indicated by a straight line. Focusing on the data D1 (β), -3π ≦ β <
The range A of π is data of two rotations (4π minutes) near the position z = 0 where the image is generated, and the data D1 of the range A
It can be seen that known two-rotation linear interpolation may be performed using (β). That is, for the data D1 (β) in the range A, an interpolation coefficient may be calculated by a known interpolation coefficient calculation method of two-rotation linear interpolation. Equation (1) is not excluded from the method of calculating the interpolation coefficient of the known two-rotation linear interpolation. Similarly,
Focusing on the data D2 (β), the range B of −π ≦ β <3π corresponds to two rotations (4) near the position z = 0 where the image is generated.
π), and it can be seen that known two-rotation linear interpolation may be performed using the data D2 (β) in the range B. That is, for the data D2 (β) in the range B,
The interpolation coefficient may be calculated by a method of calculating the interpolation coefficient of the rotation linear interpolation.

FIG. 13D shows data D1 in the range A.
FIG. 9 is an explanatory diagram of a correction coefficient for (β). −π ≦ β <
The correction coefficient for the data D1 (β) in the range of π is w1
, Data D1 in the range of −3π ≦ β <−π
Since the correction coefficient for (β) can be easily obtained by (1-w1), the data D1 in the range of −π ≦ β <π is actually obtained.
Only the correction coefficient w1 for (β) needs to be calculated.

Returning to FIG. 8, in step C3, 1 <p ≦
1.5 is determined, and if so, the process proceeds to Step C4.
Otherwise, go to step C5. In step C4, w1 (βi, γ) = {βi + π (2-1 / p)} / {2π (1-1 / p) in the range of −π (2-1 / p) <βi ≦ −π / p. )｝ (2) In the range of -π / p <βi <-π (2-1 / p) + π-2γ, w1 (βi, γ) = {βi + π (1-1 / p) + 2γ} / ｛π (1-2 / p) + 2γ｝ (3) The interpolation coefficient w1 (βi, γ) is calculated. And
Go to step R4 in FIG. The basis of the equations (2) and (3) will be described next with reference to FIG. FIG. 14 shows a case where p = 1.5, but if this is generalized, the above-mentioned step C4 is performed.

FIG. 14 is an explanatory diagram of a method of calculating an interpolation coefficient when p = 1.5. FIG. 14A shows that β = 0,
The range of the slice thickness corresponding to the data D1 (β) and D2 (β) at 2π and 4π is shown. (B) of FIG.
Indicates the position on the z-axis of the scan plane of the data D1 (β) and D2 (β) by a cosine curve. Data D1
The relative view angle of (β) is β (D1), and the data D2
The relative view angle of (β) is β (D2). -4π /
Data D1 (β) in the range of 3 (point a) ≦ β (D1) <− 2π / 3 (point b) and 2π / 3 (point b) ≦ β (D2) <4π / 3
The data D2 (β) in the range of the point (c) is the same view as the view β of the data for generating the image and is the closest to the image generation position z = 0 with the image generation position z = 0 interposed therebetween. Since there are two sets of data, it can be seen that two-rotation linear interpolation should be performed using these data. On the other hand, -2π
Data D1 (β) and π / 3-2γ (point f) ≦ β (D2) <in the range of / 3 (point b) ≦ β (D1) <− π / 3−2γ (point e)
The data D2 (β) in the range of 2π / 3 (point b) is an opposite view as a view β of data for generating an image, and the image generation position z =
Since these are the two data closest to 0, an interpolation operation (hereinafter, 1) that uses these data to obtain data of a plurality of views required for known half reconstruction.
(Referred to as rotational linear interpolation).

FIG. 14C shows data D1 (β), D
The position on the z-axis of the scan plane 2 (β) is indicated by a straight line. −4π / 3 (point a) ≦ β (D1) <− 2π / 3 (b
Data D1 (β) and 2π / 3 (point b) in range A1
≦ β (D2) <4π / 3 (point c) Data D2 in range B2
(Β) is the same view (difference is 2π) as the view β of the data for generating the image, and the image generation position z =
Since two data are closest to the image generation position z = 0 with 0 interposed therebetween, it can be seen that two-rotation linear interpolation should be performed using these data. That is, for data D1 (β) in range A1 and data D2 (β) in range B2,
What is necessary is just to calculate an interpolation coefficient by the calculation method of the interpolation coefficient of rotation linear interpolation. Equation (2) is not excluded from the calculation method of the interpolation coefficient of the two-rotation linear interpolation. On the other hand, -2π / 3 (point b) ≦ β (D
1) Data D1 in the range A2 of <−π / 3−2γ (point e)
(Β) and π / 3−2γ (point f) ≦ β (D2) <2π / 3 (b
The data D2 (β) in the range B1 of the point B) is defined as a view β of data for generating an image as an opposite view (a difference of π−
2γ), which is the two data closest to the image generation position z = 0 with the image generation position z = 0 interposed therebetween, it can be seen that linear rotation interpolation should be performed using these data. That is, the data D1 (β) in the range A2 and the data in the range B1
For the data D2 (β), the interpolation coefficient may be calculated by a known interpolation coefficient calculation method for one-rotation linear interpolation.
Equation (3) is not excluded from the calculation method of the interpolation coefficient for one-rotation linear interpolation.

FIG. 14D shows data D1 in the range A1.
FIG. 9 is an explanatory diagram of a correction coefficient for (β) and data D2 (β) in a range B2 and a correction coefficient for data D1 (β) in a range A2 and data D2 (β) in a range B1. -4
Range A of π / 3 (point a) ≦ β (D1) <− 2π / 3 (point b)
Assuming that the correction coefficient for the data D1 (β) of 1 is w1, the correction coefficient for the data D2 (β) in the range B2 of 2π / 3 (point b) ≦ β (D2) <4π / 3 (point c) Is (1
−w1), so that it is actually −4π / 3
(Point a) ≦ β (D1) <− 2π / 3 (point b) Only the correction coefficient w1 for the data D1 (β) in the range A1 may be calculated. Also, -2π / 3 (point b) ≦ β (D1) <− π / 3
When the correction coefficient for the data D1 (β) in the range A2 of −2γ (point e) is w1, π / 3−2γ (point f) ≦
Data D2 in the range B1 of β (D2) <2π / 3 (point b)
Since the correction coefficient for (β) can be easily obtained by (1−w1), actually −2π / 3 (point b) ≦ β (D1) <−
Only the correction coefficient w1 for the data D1 (β) in the range A2 of π / 3−2γ (point e) needs to be calculated.

Returning to FIG. 8, in step C5, 1.5 <
It is determined whether or not p ≦ 3, and if so, the process proceeds to Step C6,
Otherwise, go to step C7. In step C6, w1 (βi, γ) = {βi + π (2-1 / p)} / {2π (1-1 / p) in the range of −π (2-1 / p) <βi ≦ −π / p. )｝ (4) In the range of -π / p <βi <π / p, the interpolation coefficient w1 (βi is given by w1 (βi, γ) = (π / p-βi) / (2π / p) (5) , Γ). And
Go to step R4 in FIG. The basis of the equations (4) and (5) will be described next with reference to FIG. FIG. 15 shows a case where p = 2, but if this is generalized, the above-mentioned step C6 is performed.

FIG. 15 is an explanatory diagram of a method of calculating an interpolation coefficient when p = 2. FIG. 15A shows that β = 0, 2π
2 shows the range of the slice thickness corresponding to the data D1 (β) and D2 (β). FIG. 15B shows the positions on the z-axis of the scan plane of the data D1 (β) and D2 (β) by a cosine curve. The relative view angle of the data D1 (β) is β (D1), and the relative view angle of the data D2 (β) is β (D2). -3π / 2 (point a) ≦ β (D1)
Data D1 (β) and π / 2 in the range of <−π / 2 (point b)
Data D in the range of (point b) ≦ β (D2) <3π / 2 (point c)
2 (β) is a view β of data for generating an image.
Since the two views are the same view and are two pieces of data that are closest to the image generation position z = 0 with the image generation position z = 0 interposed therebetween, it is understood that two-rotation linear interpolation may be performed using these data. On the other hand, -π / 2 (point b) ≦ β (D1) <π /
Data D1 (β) in the range of 2 (point c) and −π / 2 (a
Point) ≦ β (D2) <π / 2 (point b)
(Β) are two data that are the same view as the view β of data for generating an image and that are closest to the image generation position z = 0 with the image generation position z = 0 interposed therebetween. It can be seen that two-rotation linear interpolation should be performed using

FIG. 15C shows data D1 (β), D
The position on the z-axis of the scan plane 2 (β) is indicated by a straight line. −3π / 2 (point a) ≦ β (D1) <− π / 2 (b
Data D1 (β) in the range A1 and π / 2 (point b) ≦
Data D2 in the range B2 of β (D2) <3π / 2 (point c)
(Β) is the same view (the difference is 0) as the view β of the data for generating the image, and the image generation position z = 0.
, And the two data closest to the image generation position z = 0, it can be seen that it is sufficient to perform two-rotation linear interpolation using these data. That is, the data D in the range A1
For 1 (β) and the data D2 (β) in the range B2, the interpolation coefficient may be calculated by an interpolation coefficient calculation method of two-rotation linear interpolation. Equation (4) is not excluded from the calculation method of the interpolation coefficient of the two-rotation linear interpolation. On the other hand, -π / 2 (point b) ≦ β (D1) <
The data D1 (β) in the range A2 of π / 2 (point c) and −π /
The data D2 (β) in the range B1 of 2 (point a) ≦ β (D2) <π / 2 (point b) is the same view (the difference is 2π) as the view β of the data for generating the image. Since the two data are closest to the image generation position z = 0 with respect to the image generation position z = 0, it can be seen that two-rotation linear interpolation should be performed using these data. That is, the interpolation coefficient may be calculated for the data D1 (β) in the range A2 and the data D2 (β) in the range B1 by the interpolation coefficient calculation method of the two-rotation linear interpolation. The expression (5) is not excluded from the calculation method of the interpolation coefficient of the two-rotation linear interpolation.

FIG. 15D shows data D1 in the range A1.
FIG. 9 is an explanatory diagram of a correction coefficient for (β) and data D2 (β) in a range B2 and a correction coefficient for data D1 (β) in a range A2 and data D2 (β) in a range B1. -3
Range A1 of π / 2 (point a) ≦ β (D1) <− π / 2 (point b)
Assuming that the correction coefficient for the data D1 (β) is w1, the correction coefficient for the data D2 (β) in the range B2 of π / 2 (point b) ≦ β (D2) <3π / 2 (point c) is (1-w
Since it is easily obtained by 1), actually -3π / 2 (a
Point) ≦ β (D1) <− π / 2 (point b) Data D in range A1
Only the correction coefficient w1 for 1 (β) needs to be calculated. -Π / 2 (point b) ≦ β (D1) <π / 2 (point c)
The correction coefficient for the data D1 (β) in the range A2
Assuming that 1, −π / 2 (point a) ≦ β (D2) <π / 2 (b
Since the correction coefficient for the data D2 (β) in the range B1 of (point) can be easily obtained by (1−w1), actually −π
Only the correction coefficient w1 for the data D1 (β) in the range A2 of / 2 (point b) ≦ β (D1) <π / 2 (point c) may be calculated.

Returning to FIG. 8, in step C7, -π (2
In the range of −1 / p) + π−2γ <βi ≦ −π / p, w1 (βi, γ) = {βi + π (1-1 / p) +2
γ} / {π (1-2 / p) + 2γ} (6) −π /
In the range of p <βi <π / p, w1 (βi, γ) =
(Π / p−βi) / (2π / p) (7) is used to calculate the interpolation coefficient w1 (βi, γ). Then, the process proceeds to step R4 in FIG. The basis of the equations (6) and (7) will be described next with reference to FIG. FIG. 16 shows the case where p = 3. If this is generalized, step C7 is obtained.

FIG. 16 is an explanatory diagram of a method of calculating an interpolation coefficient when p = 3. FIG. 16A shows that β = 0, 2π
2 shows the range of the slice thickness corresponding to the data D1 (β) and D2 (β). FIG. 16B shows the positions on the z-axis of the scan plane of the data D1 (β) and D2 (β) by a cosine curve. The relative view angle of the data D1 (β) is β (D1), and the relative view angle of the data D2 (β) is β (D2). -2π / 3-2γ (point a) ≦
Data D1 (β) in the range of β (D1) <− π / 3 (point b) and data D2 (β in the range of π / 3 (point b) ≦ β (D2) <2π / 3-2γ (point c) β) is an opposite view as a view β of data for generating an image, and an image generation position z
Since the two data are closest to the image generation position z = 0 with respect to = 0, it can be seen that one-rotation linear interpolation should be performed using these data. On the other hand, -π / 3 (point b) ≦
Data D1 (β) in the range of β (D1) <π / 3 (point e) and −
The data D2 (β) in the range of π / 3 (point f) ≦ β (D2) <π / 3 (point b) is the same view as the view β of the data for generating the image and the image generation position Since the two data are closest to the image generation position z = 0 with z = 0 interposed therebetween, it can be seen that two-rotation linear interpolation should be performed using these data.

FIG. 16C shows data D1 (β), D
The position on the z-axis of the scan plane 2 (β) is indicated by a straight line. −2π / 3−2γ (point a) ≦ β (D1) <− π / 3
The data D1 (β) and π / 3 (b) in the range A1 of (point b)
Point) ≦ β (D2) <2π / 3−2γ (point c) The data D2 (β) in the range B2 is an opposite view (difference of π−2γ) as a view β of data for generating an image. There are two pieces of data closest to the image generation position z = 0 with the image generation position z = 0 interposed therebetween.
It can be seen that rotational linear interpolation should be performed. That is, the data D1 (β) of the range A1 and the data D2 (β) of the range B2
With respect to the above, the interpolation coefficient may be calculated by a method for calculating the interpolation coefficient of one-rotation linear interpolation. Equation (6) is not excluded from the calculation method of the interpolation coefficient for one-rotation linear interpolation. On the other hand, -π / 3 (b
Point) ≦ β (D1) <π / 3 (point e) Data D1 in range A2
(Β) and data D2 (β) in the range B1 of −π / 3 (point f) ≦ β (D2) <π / 3 (point b) are the same view as the view β of the data for generating the image. (The difference is 0), and the two data are closest to the image generation position z = 0 with the image generation position z = 0 interposed therebetween. Therefore, it is understood that two-rotation linear interpolation may be performed using these data. . That is, the data D1 (β) in the range A2 and the data D1 in the range B1
For 2 (β), the interpolation coefficient may be calculated by a known interpolation coefficient calculation method of two-rotation linear interpolation. Equation (7) is 2
It does not deviate from the method of calculating the interpolation coefficient for rotational linear interpolation.

FIG. 16D shows data D1 in the range A1.
FIG. 9 is an explanatory diagram of a correction coefficient for (β) and data D2 (β) in a range B2 and a correction coefficient for data D1 (β) in a range A2 and data D2 (β) in a range B1. -2
When the correction coefficient for the data D1 (β) in the range A1 of π / 3−2γ (point a) ≦ β (D1) <− π / 3 (point b) is w1, π / 3 (point b) ≦ β (D2) <2π / 3-2γ
Since the correction coefficient for the data D2 (β) in the range B2 of the (point c) can be easily obtained by (1−w1), actually −2π / 3−2γ (a point) ≦ β (D1) <− π / 3 (point b)
Correction coefficient w1 for data D1 (β) in range A1
You only need to calculate -Π / 3 (point b) ≦ β (D
1) When the correction coefficient for the data D1 (β) in the range A2 of <π / 3 (point e) is w1, −π / 3 (point f) ≦
Data D2 (β) in the range B1 of β (D2) <π / 3 (point b)
Can be easily obtained from (1-w1), so that -π / 3 (point b) ≦ β (D1) <π / 3 (e
Only the correction coefficient w1 for the data D1 (β) in the range A2 of (point) need be calculated.

FIG. 9 shows a process of reading an interpolation coefficient set (FIG. 7).
It is a flowchart of step R3). Step F
At 1, it is determined whether or not p = 1.5, and if so, the process proceeds to step F2; otherwise, the process proceeds to step F3. In step F2, the correction coefficient w1 previously calculated and stored with p = 1.5 is read out from the above equations (2) and (3).
Then, the process proceeds to step R4 in FIG. In step F3, it is determined whether or not p = 2. If so, the process proceeds to step F4; otherwise, the process proceeds to step F5. Step F4
Then, the correction coefficient w1 calculated and stored in advance as p = 2 in the above equations (4) and (5) is read. And FIG.
To step R4. In step F5, the above (6)
In equation (7), the correction coefficient w1 calculated and stored in advance as p = 3 is read. Then, the process proceeds to step R4 in FIG.

FIG. 10 is a flowchart of a data calculation process by interpolation calculation. In step G1, p = 0.
It is determined whether it is 5 or not, and if so, the process proceeds to Step G2; otherwise, the process proceeds to Step G4. In step G2,-
In the range of 4π ≦ βi <0, D (βi, γ) = （D1 (βi, γ) + D2 (βi + 4π,
γ)｝ / 2 to calculate average data D (βi, γ). As can be seen from FIG. 12, the data D1 (β, γ) and D2 (β + 4
π, γ) are data of the same scan plane and the same view (difference is 4π) as a view of data for generating an image, so that D1 (β, γ) = D2 (β + 4)
π, γ). Therefore, D1 (β, γ) and D2 (β +
4π, γ) can be added and averaged, whereby the SN ratio (Sig
nal to Noise Ratio). In step G3, using the interpolation coefficient w1 obtained in step C2 and the average data D (βi, γ), data of a plurality of views DF ( βi, γ). This data D
F (βi, γ) is consequently the data D in the range A in FIG.
1 (β, γ) and the data D2 (β, γ) in the range B. Then, the process proceeds to step R5 in FIG.

In step G4, it is determined whether or not p <1, and if so, the process proceeds to step G5, and if not, the process proceeds to step G6. In Step G5, Step C2
Using the interpolation coefficient w1 obtained in the above and the data D1 (β, γ) in the range A in FIGS. 12 and 13, the data DF () of a plurality of views necessary for generating an image by a known two-rotation linear interpolation operation βi, γ). Then, the process proceeds to step R5 in FIG.

In step G6, it is determined whether or not p = 1, and if so, the process proceeds to step G7.
Proceed to Step G9 of Step 1. In step G7, -3π ≦
In the range of βi <π, D (βi, γ) = ｛D1 (βi, γ) + D2 (βi + 2π,
γ)｝ / 2 to calculate average data D (βi, γ). As can be seen from FIG. 13, data D1 (β, γ) and D2 (β + 2
π, γ) is data of the same scan plane and the same view (difference is 2π) as a view of data for generating an image, so D1 (β, γ) = D2 (β + 2)
π, γ). Therefore, D1 (β, γ) and D2 (β +
2π, γ) can be added and averaged, thereby improving the SN ratio. In step G8, using the interpolation coefficient w1 obtained in step C2 and the average data D (βi, γ), data of a plurality of views DF ( βi, γ). The data DF (βi, γ) is based on the data D1 (β, γ) in the range A and the data D2 (β, γ) in the range B in FIG. And FIG.
Go to step R5.

In step G9 of FIG. 11, 1 <p ≦ 1.
It is determined whether it is 5 or not, and if so, the process proceeds to step G10; otherwise, the process proceeds to step G11. In step G10, the interpolation coefficient w1 obtained in step C4 or step F2 and the data D1 in the range A1 in FIG.
Using (β, γ) and the data D2 (β, γ) in the range B2, data DF (βi, γ) of a plurality of views required to generate an image is calculated by a two-rotation linear interpolation operation. That is, in the range of -π (2-1 / p) <βi ≦ -π / p, DF (βi, γ) = w1 (βi, γ) · D1 (βi, γ) + (1-w1 (βi, γ)) · D2 (βi + 2π, γ)…
(8) Step C4 or step F
14 and the data D1 (β, γ) in the range A2 and the data D2 (β, γ) in the range B1 in FIG.
γ), data DF (βi, γ) of a plurality of views required to generate an image by one-rotation linear interpolation operation
Is calculated. That is, −π / p <βi <−π (2-1
/ P) + π−2γ, DF (βi, γ) = w1 (βi, γ) · D1 (βi, γ) + (1−w1 (βi, γ)) · D2 (βi + π−2γ, −γ )… (9). Then, the process proceeds to step R5 in FIG.

In step G11, it is determined whether or not 1.5 <p <3. If so, the flow proceeds to step G12, and if not, the flow proceeds to step G13. In step G12, the interpolation coefficient w1 obtained in step C6 or step F4 and the data D1 in the ranges A1 and A2 in FIG.
(Β, γ) and data D2 (β,
γ), data DF (βi, γ) of a plurality of views necessary for generating an image by a two-rotation linear interpolation operation
Is calculated. Then, the process proceeds to step R5 in FIG.

In step G13, the interpolation coefficient w1 obtained in step C7 or step F5 and the interpolation coefficient w1 shown in FIG.
The data D1 (β, γ) of the range A1 and the data D2 (β, γ) of the range B2 are used to perform a one-rotation linear interpolation operation, and the data DF (βi, β) of a plurality of views necessary for generating an image. γ) is calculated. That is, -π (2-1)
/ P) + π-2γ <βi ≦ −π / p, DF (βi, γ) = w1 (βi, γ) · D1 (βi, γ) + (1-w
1 (βi, γ)) · D2 (βi + π−2γ, γ). Step C7 or step F
5 and the data D1 (β, γ) of the range A2 and the data D2 (β, γ) of the range B1 in FIG.
γ), data DF (βi, γ) of a plurality of views necessary for generating an image by a two-rotation linear interpolation operation
Is calculated. That is, in the range of -π / p <βi <π / p, DF (βi, γ) = w1 (βi, γ) · D1 (βi, γ) + (1-w
1 (βi, γ)) · D2 (βi + 2π, −γ). Then, the process proceeds to step R5 in FIG.

According to the X-ray CT apparatus 100 described above, data can be collected by helical scanning using a two-stage detector array, and an image can be suitably generated.

Second Embodiment The present invention can be similarly applied to the case where the number N of stages of the detector array is three or more. FIG. 17 shows a conceptual diagram of a data range and interpolation coefficients used when N = 3 and p = 3. FIG. 17A shows the range of the slice thickness corresponding to the data D1 (β), D2 (β), and D3 (β) at β = 0 and 2π. FIG. 17B shows data D
The positions of the scan planes 1 (β), D2 (β), and D3 (β) on the z-axis are indicated by cosine curves. Data D1
The relative view angle of (β) is β (D1), and the data D2
The relative view angle of (β) is β (D2), and the data D3
The relative view angle of (β) is β (D3). -4π /
Data D1 (β) in the range of 3 (point a) ≦ β (D1) <− 2π / 3 (point b) and 2π / 3 (point b) ≦ β (D3) <4π / 3
The data D3 (β) in the range of the point (c) is the same view as the view β of the data for generating an image, and is the closest view to the image generation position z = 0 across the image generation position z = 0. Since there are two sets of data, it can be seen that two-rotation linear interpolation should be performed using these data. Next, -2π
Data D1 in the range of / 3 (point b) ≦ β (D1) <0 (point e)
(Β) and the data D2 (β) in the range of −2π / 3 (point a) ≦ β (D2) <0 (point f) are the same view as the view β of the data for generating the image. Image generation position z
Since the two data are closest to the image generation position z = 0 with respect to = 0, it can be seen that two-rotation linear interpolation should be performed using these data. Further, 0 (point f) ≦ β (D
2) Data D2 (β) and 0 in the range of <2π / 3 (point c)
(Point g) ≦ β (D3) <2π / 3 (point b)
3 (β) is a view β of data for generating an image.
Since the two views are the same view and are two pieces of data that are closest to the image generation position z = 0 with the image generation position z = 0 interposed therebetween, it is understood that two-rotation linear interpolation may be performed using these data.

FIG. 17C shows data D1 (β), D
The positions on the z-axis of the scan planes 2 (β) and D3 (β) are indicated by straight lines. −4π / 3 (point a) ≦ β (D1) <−
Data D1 (β) and 2π in the range A1 of 2π / 3 (point b)
The data D3 (β) in the range C2 of / 3 (point b) ≦ β (D3) <4π / 3 (point c) is the same view (difference is 2π) as the view β of the data for generating the image. Since the two data are closest to the image generation position z = 0 with respect to the image generation position z = 0, it can be seen that two-rotation linear interpolation may be performed using these data. That is, the range A
The interpolation coefficient may be calculated for the data D1 (β) of 1 and the data D3 (β) of the range C2 by the interpolation coefficient calculation method of the two-rotation linear interpolation. Next, −2π / 3 (point b) ≦ β (D
1) Data D1 (β) and −2π in the range A2 of <0 (point e)
The data D2 (β) in the range B1 of / 3 (point a) ≦ β (D2) <0 (point f) is the same view (the difference is 0) as the view β of the data for generating the image. Image generation position z
Since the two data are closest to the image generation position z = 0 with respect to = 0, it can be seen that two-rotation linear interpolation should be performed using these data. That is, for the data D1 (β) in the range A2 and the data D2 (β) in the range B1,
What is necessary is just to calculate an interpolation coefficient by the calculation method of the interpolation coefficient of two rotation linear interpolation. Further, 0 (point f) ≦ β (D2) <2π / 3
Data D2 (β) in the range B2 (point c) and 0 (point g) ≦
Data D3 in the range C1 of β (D3) <2π / 3 (point b)
(Β) are two data that are the same view as the view β of data for generating an image and that are closest to the image generation position z = 0 with the image generation position z = 0 interposed therebetween. It can be seen that two-rotation linear interpolation should be performed using That is, data D2 (β) in the range B2
And the data D3 (β) in the range C1 may be calculated by an interpolation coefficient calculation method of two-rotation linear interpolation.

FIG. 17D shows data D1 in the range A1.
(Β) and correction coefficient for data D3 (β) in range C2, data D1 (β) for range A2 and data D for range B1
Correction coefficient for 2 (β) and data D in range B2
FIG. 4 is an explanatory diagram of correction coefficients for 2 (β) and data D3 (β) in a range C1.

[0055]

According to the X-ray CT apparatus of the present invention, data can be collected by helical scan using a multi-stage detector array, and an image can be suitably generated. And
Image quality can be improved.

[Brief description of the drawings]

FIG. 1 is a block diagram of an X-ray CT apparatus according to a first embodiment.

FIG. 2 is a schematic diagram of a two-stage detector array of the X-ray CT apparatus of FIG.

FIG. 3 is an explanatory diagram of a situation in which scanning is performed by the X-ray CT apparatus of FIG. 1;

FIG. 4 is an explanatory diagram of an absolute view angle, a relative view angle, and an opposite view.

FIG. 5 is an explanatory diagram of a relative view angle and a position on a linear movement axis.

FIG. 6 is a flowchart of a helical scan process.

FIG. 7 is a flowchart of an image reconstruction process.

FIG. 8 is a flowchart of a calculation process of an interpolation coefficient set.

FIG. 9 is a flowchart of an interpolation coefficient set reading process.

FIG. 10 is a flowchart of the first half of a data calculation process by an interpolation calculation.

FIG. 11 is a flowchart of the latter half of data calculation processing by interpolation calculation.

FIG. 12 is an explanatory diagram of a method of calculating an interpolation coefficient when the number of stages of the detector array is N = 2 and the helical pitch is p = 0.5.

FIG. 13 is an explanatory diagram of a method of calculating an interpolation coefficient when the number of stages of the detector array is N = 2 and the helical pitch is p = 1.

FIG. 14 is an explanatory diagram of a calculation method of an interpolation coefficient when the number of stages of the detector array is N = 2 and the helical pitch is p = 1.5.

FIG. 15 is an explanatory diagram of a method of calculating an interpolation coefficient when the number of stages of the detector array is N = 2 and the helical pitch is p = 2.

FIG. 16 is an explanatory diagram of a method of calculating an interpolation coefficient when the number of stages of the detector array is N = 2 and the helical pitch is p = 3.

FIG. 17 is an explanatory diagram of a method of calculating an interpolation coefficient when the number of stages of the detector array is N = 3 and the helical pitch is p = 3.

[Explanation of symbols]

Reference Signs List 100 X-ray CT apparatus 3 Central processing unit 20 Scanning gantry 30 X-ray tube 60 Two-stage detector array 61 First detector array 62 Second detector array Xr Fan-shaped X-ray beam th Fan-shaped X-ray beam thickness d Linear movement distance per rotation p Helical pitch

Claims (2)

(57) [Claims] An object to be scanned is relatively straight along an axis in the juxtaposition direction with respect to a multi-stage detector array in which a plurality of X-ray detectors are arranged in a line in a row. Data acquisition means for acquiring data while moving and rotating the X-ray tube around the scanning object; and data of a plurality of views necessary to generate an image at one position on the linear movement axis. What is claimed is: 1. An X-ray CT apparatus comprising: a data calculation unit that calculates by interpolation from data collected in the vicinity of a generation position; and an image generation unit that generates an image from the data of the plurality of views calculated by the data calculation unit. The data calculation means collects, at each detector array, two pieces of data that are the same view data or the opposite view data of each of the plurality of views and are closest to the image generation position. While selecting from the data
The number of stages of the detector array of the multi-stage detector array is N,
Let the thickness of the fan-shaped X-ray beam incident on each detector array be th
When the distance linearly moved relative to each rotation of the X-ray tube or the multi-stage detector array is d, and p = d / th is a helical pitch, the number of stages N and the helical pitch p
An X-ray CT apparatus, wherein an interpolation coefficient is determined based on the above, and an interpolation operation is performed using the interpolation coefficient and the selected data to calculate data for generating an image. 2. A multi-stage detector array in which a large number of X-ray detectors are arranged in a row in two or more stages. Data acquisition means for acquiring data while moving and rotating the X-ray tube around the scanning object; and data of a plurality of views necessary to generate an image at one position on the linear movement axis. What is claimed is: 1. An X-ray CT apparatus comprising: a data calculation unit that calculates by interpolation from data collected in the vicinity of a generation position; and an image generation unit that generates an image from the data of the plurality of views calculated by the data calculation unit. Let the thickness of the fan-shaped X-ray beam incident on each detector array be th, and let d be the distance linearly moved relative to each rotation of the X-ray tube or multistage detector array, and p = d /
When th is a helical pitch, the data collection unit allows the operator to select any one of a plurality of preset ps, collects data at the selected helical pitch p, and Calculates and stores in advance interpolation coefficients for each of a plurality of preset p's, and stores data of the same view or data of the opposite view as each of the plurality of views, which is closest to the image generation position. An X-ray CT apparatus comprising: selecting one data from data collected by each detector array; performing an interpolation operation using the stored interpolation coefficient; and calculating data for generating an image.