JP5317479B2 - X-ray CT system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an X-ray CT apparatus capable of reducing the displacement of tomographic images with different X-ray tube voltages. <P>SOLUTION: The X-ray CT apparatus 100 includes an X-ray tube 21 for irradiating a subject with X rays having a first energy spectrum and X rays with a second energy spectrum; an X-ray data collecting part 24 for collecting X-ray projection data; an image reconstruction part 34 for reconstructing a first tomographic image and a second tomographic image; a tomographic image displacement correction part 37 for finding the common center of gravity to the first tomographic image and the second tomographic image, finding the displacement between the first tomographic image and the second tomographic image based on the comparison with an object for comparison set in the direction radially extending from the center of gravity, and correcting at least either the first tomographic image or the second tomographic image based on the quantity of displacement; and a dual energy image reconstruction means 35 for reconstructing a tomographic image of the information depending on the X-ray tube voltage based on the positioned first tomographic image and second tomographic image. <P>COPYRIGHT: (C)2009,JPO&amp;INPIT

Description

  The present invention optimizes the spatial resolution and image noise of a two-dimensional distribution tomogram of X-ray tube voltage-dependent information in an X-ray absorption coefficient related to the distribution of atoms, so-called dual energy scan tomogram. The present invention relates to X-ray CT apparatus technology.

  If you want to take a tomogram of material distribution, so-called Dual Energy Scan, using two tomograms of X-ray tube voltage, a tomogram dual in which artifacts due to misalignment represent X-ray tube voltage-dependent information There was a problem of appearing in tomographic images of energy imaging. For example, when imaging a lung field, if a low X-ray tube voltage is exhaust and a high X-ray tube voltage is expiration, a low X-ray tube voltage tomogram and a high X-ray tube voltage tomogram are positioned. Slip. In this state, when a weighted addition process is performed on a tomographic image with a low X-ray tube voltage and a tomographic image with a high X-ray tube voltage, a mis-registration artifact occurs.

Patent Document 1 discloses a technique for matching or aligning a low energy image and a high energy image in dual energy imaging in X-ray imaging using flat panel technology.
JP 2003-244542 A

  In the dual energy imaging in the X-ray CT apparatus, in order to obtain a tomographic image, the influence of the positional deviation on the movement of the organ of the subject becomes large, and a technique for highly accurate positional deviation correction has been desired.

  Accordingly, an object of the present invention is to use a two-dimensional distribution tomogram showing X-ray tube voltage dependency information in an X-ray absorption coefficient related to the distribution of atoms, that is, a so-called dual energy imaging tomogram by an X-ray CT apparatus. An object of the present invention is to provide an X-ray CT apparatus that reduces a decrease in spatial resolution.

An X-ray CT apparatus according to a first aspect includes an X-ray tube that irradiates a subject with X-rays having a first energy spectrum and X-rays having a second energy spectrum different from the first energy spectrum, An X-ray data collection unit that collects the X-ray projection data of the first energy spectrum and the X-ray projection data of the second energy spectrum irradiated to the subject; the X-ray projection data of the first energy spectrum; Based on the X-ray projection data of the second energy spectrum, an image reconstruction unit that reconstructs the first tomogram and the second tomogram, and a center of gravity common to the first tomogram and the second tomogram are obtained. A displacement amount between the first tomographic image and the second tomographic image is obtained by comparison with a comparison target set in a direction extending radially from the center of gravity, and the first tomographic image and A tomographic image displacement correction unit that performs displacement correction based on the displacement amount on at least one of the second tomographic images, and an atomic distribution based on the aligned first tomographic image and the second tomographic image Dual energy image reconstruction means for reconstructing a tomographic image of X-ray tube voltage-dependent information related to the above. In a second aspect, the X-ray CT apparatus according to the second aspect is the X-ray CT apparatus according to the first aspect, wherein the tomographic image position deviation correction unit sets the one point as a region having a high CT value of the subject. Is to select from.
An X-ray CT apparatus according to a third aspect is the X-ray CT apparatus according to the second aspect, wherein the region having a high CT value of the subject is a bone region.
According to a fourth aspect of the X-ray CT apparatus, in any one of the first to third aspects, the tomographic image position misalignment correction unit applies the tomographic image to each of the first tomographic image and the second tomographic image. Is divided into a plurality of sectors extending radially from the center of gravity, and the first tomographic image and the first tomographic image are compared by comparing the sector-shaped region of the first tomographic image with the sector-shaped region of the second tomographic image. The amount of positional deviation from the two tomographic images is measured.
According to a fifth aspect of the X-ray CT apparatus, in the fourth aspect, the positional deviation amount is obtained by a similarity ratio between the sectoral area of the first tomographic image and the sectoral area of the second tomographic image. Is.
An X-ray CT apparatus according to a sixth aspect is the X-ray CT apparatus according to the fourth aspect, wherein the positional deviation amount is obtained by a correlation operation between the sector area of the first tomogram and the sector area of the second tomogram. It is.
In the fourth or fifth aspect, the X-ray CT apparatus according to a seventh aspect is configured such that the tomographic image position correction unit overlaps the sector regions with respect to a portion where each sector region is in contact with an adjacent sector region, In the overlapped portion, the weighted addition process is performed so that the sector regions are continuously joined.
In an X-ray CT apparatus according to an eighth aspect, in any one of the first to third aspects, the tomographic image position deviation correction unit radiates from the center of gravity in each of the first tomographic image and the second tomographic image. And a pixel value on the line segment of the first tomographic image is compared with a pixel value on the line segment of the second tomographic image. The amount of positional deviation from the second tomographic image is obtained.
The X-ray CT apparatus according to a ninth aspect is the X-ray CT apparatus according to the eighth aspect, wherein the positional deviation amount is a pixel value on the line segment of the first tomographic image and a pixel value on the line segment of the second tomographic image. It is obtained by the correlation calculation.
In an X-ray CT apparatus according to a tenth aspect, in the sixth or ninth aspect, the positional deviation amount is an enlargement / reduction ratio calculated using a half-value width of a profile obtained as a result of the correlation calculation. Is.
In an X-ray CT apparatus according to an eleventh aspect, in the first to tenth aspects, the tomographic image position shift correction unit displays the first tomographic image and the second tomographic image in a superimposed manner for comparison. It is.
In an X-ray CT apparatus according to a twelfth aspect, in the first to tenth aspects, the tomographic image positional deviation correction unit is based on the positional deviation amount based on at least one of the first tomographic image and the second tomographic image. Coordinate transformation is performed.
An X-ray CT apparatus according to a thirteenth aspect is the twelfth aspect, wherein the coordinate transformation is affine transformation or multi-order coordinate transformation.

  According to the X-ray CT apparatus of the present invention, even if the subject moves due to breathing or pulsation, the dual resolution imaging reduces the reduction in spatial resolution due to tomographic image misalignment. There is an effect that a line CT apparatus can be realized.

<Overall configuration of X-ray CT apparatus>
FIG. 1 is a configuration block diagram of an X-ray CT apparatus 100 according to an embodiment of the present invention. The X-ray CT apparatus 100 includes an operation console 1, an imaging table 10, and a scanning gantry 20.

  The operation console 1 includes an input device 2 such as a keyboard or a mouse that receives input from the operator, a central processing unit 3 that executes preprocessing, image reconstruction processing, postprocessing, and the like, and X-ray detection collected by the scanning gantry 20 And a data collection buffer 5 for collecting the vessel data. Further, the operation console 1 includes a monitor 6 that displays a tomographic image reconstructed from projection data obtained by preprocessing X-ray detector data, a program, X-ray detector data, projection data, and X-ray tomography. And a storage device 7 for storing an image. The photographing condition is input from the input device 2 and stored in the storage device 7. The imaging table 10 includes a cradle 12 on which a subject is placed and taken in and out of the opening of the scanning gantry 20. The cradle 12 is moved up and down and moved linearly by a motor built in the imaging table 10.

  The scanning gantry 20 includes an X-ray tube 21, an X-ray controller 22, a collimator 23, a beam forming X-ray filter 28, a multi-row X-ray detector 24, and a data acquisition device (DAS: Data Acquisition System) 25. It has. Further, the scanning gantry 20 includes a rotating unit controller 26 that controls the X-ray tube 21 rotating around the body axis of the subject, and a control controller 29 that exchanges control signals and the like with the operation console 1 and the imaging table 10. It is equipped with. The beam forming X-ray filter 28 has the thinnest filter thickness in the X-ray direction toward the center of rotation, which is the imaging center, and the filter thickness increases toward the periphery so that X-rays can be absorbed more. X-ray filter. For this reason, exposure of the body surface of the subject having a cross-sectional shape close to a circle or an ellipse can be reduced.

The central processing unit 3 includes a preprocessing unit 31, a beam hardening processing unit 33, an image reconstruction unit 34, a dual energy image reconstruction unit 35, and a tomographic image position deviation correction unit 37.
The pre-processing unit 31 corrects non-uniform sensitivity between channels for the raw data collected by the data collection device 25, and an extreme signal intensity drop or signal due to an X-ray strong absorber, mainly a metal part. Pre-processing such as X-ray dose correction for correcting omission is executed.

  The beam hardening processing unit 33 corrects the beam hardening of the projection data. Beam hardening is a phenomenon in which X-ray absorption changes depending on the transmission thickness even with the same material, and the CT value (brightness) on the CT image changes. In particular, the energy distribution of the radiation that has passed through the subject is on the high energy side. It is biased to. For this reason, beam hardening is corrected with respect to the slice direction and the channel direction of the projection data.

  The image reconstruction unit 34 receives the projection data preprocessed by the preprocessing unit 31, and reconstructs an image based on the projection data. The projection data is subjected to Fast Fourier Transform (FFT) for transforming into the frequency domain, and the reconstruction function Kernel (j) is superimposed on the projection data, and inverse Fourier transform is performed. Then, the image reconstruction unit 34 performs a three-dimensional backprojection process on the projection data obtained by superimposing the reconstruction function Kernel (j), and obtains a tomographic image (for each body axis direction (Z direction) of the subject HB). xy plane). The image reconstruction unit 34 stores this tomographic image in the storage device 7.

The dual energy image reconstruction unit 35 reconstructs, from the projection data or tomographic image, a two-dimensional distribution tomographic image of so-called dual energy imaging tomographic image of X-ray tube voltage-dependent information related to the atom distribution.
The tomographic image position correction unit 37 corrects the positional shift of each tomographic image due to a low tube voltage and a high tube voltage in order to optimize a tomographic image of dual energy imaging. Details of the tomographic image position correction unit 37 will be described later.

<Operation flowchart of X-ray CT apparatus>
FIG. 2 is a flowchart showing an outline of the operation of the X-ray CT apparatus of the present embodiment.
In step P1, the subject is placed on the cradle 12 and aligned. Here, the subject placed on the cradle 12 aligns the slice light center position of the scanning gantry 20 with the reference point of each part. Then, a scout image (also called a scanogram or an X-ray fluoroscopic image) is collected. In scout imaging, the X-ray tube 21 and the multi-row X-ray detector 24 are fixed, and the data collection operation of X-ray detector data is performed while the cradle 12 is moved linearly. Here, the scout image is usually taken at a view angle position of 0 degrees and 90 degrees. Depending on the part, for example, the head may be a 90-degree scout image only. The right side in FIG. 2 is an example of a scout image 41 obtained by photographing the vicinity of the chest at 0 degrees. From this scout image 41, it is possible to plan the photographing position of the tomographic image.

  In step P2, the photographing condition is set while displaying the position and size of the tomographic image to be photographed on the scout image 41. The dotted line shown in the scout image 41 is the position of the tomographic image. In the present embodiment, a plurality of scan patterns such as a conventional scan, a helical scan, a variable pitch helical scan, and a helical shuttle scan are provided. The conventional scan is a scan method in which the X-ray tube 21 and the multi-row X-ray detector 24 are rotated to acquire X-ray projection data every time the cradle 12 is moved at a predetermined interval in the z-axis direction. The helical scan is an imaging method for collecting X-ray projection data by moving the cradle 12 at a constant speed while an X-ray data collection system including the X-ray tube 21 and the multi-row X-ray detector 24 rotates. The variable pitch helical scan collects X-ray projection data by changing the speed of the cradle 12 while rotating the X-ray data acquisition system including the X-ray tube 21 and the multi-row X-ray detector 24 as in the helical scan. It is a shooting method. In the helical shuttle scan, as in the helical scan, the cradle 12 is accelerated and decelerated while rotating the X-ray data acquisition system including the X-ray tube 21 and the multi-row X-ray detector 24, and the positive direction of the z axis or z This is a scanning method for collecting X-ray projection data by reciprocating in the negative direction of the axis. When these plural radiographs are set, X-ray dose information as a whole is displayed.

  In setting the tomographic image capturing conditions, the exposure of the subject can be optimized by using the automatic exposure mechanism of the X-ray CT apparatus 100. In addition, in this tomographic imaging condition setting, for the so-called dual energy imaging tomographic imaging, the X-ray tube 21 has a low X-ray tube voltage, for example, 80 kV, and a high X-ray tube voltage, for example, 140 kV. You can set conditions. Further, in the automatic exposure mechanism in dual energy imaging, the imaging condition of the low X-ray tube voltage and the high X are set so that the noise index value of the final image of the dual energy tomogram is substantially equal to the set noise index value. The imaging conditions of the tube voltage can be determined. At this time, it is X-ray exposure that the imaging conditions of the respective X-ray tube voltages are determined so that the image noise of the tomographic image of the low X-ray tube voltage and the image noise of the tomographic image of the high X-ray tube voltage are substantially equal. It is preferable from the viewpoint of optimization.

  In step P3 to step P9, tomographic imaging is performed. In step P3, X-ray data collection is performed. Here, when collecting data by helical scanning, the X-ray tube 21 and the multi-row X-ray detector 24 are rotated around the subject, and the cradle 12 on the imaging table 10 is moved linearly, Data collection operation of X-ray detector data is performed. Then, the X-ray detector data D0 (view, j, i) (j = 1 to ROW, i = 1 to CH) represented by the view angle view, the detector row number j, and the channel number i is z direction. Coordinate position Ztable (view) is added. As described above, in the helical scan, X-ray detector data collection within a constant speed range is performed. This z-direction coordinate position may be added to the X-ray projection data (X-ray detector data), or may be used in association with the X-ray projection data as a separate file. The z-direction coordinate position information is used when reconstructing the three-dimensional image of the X-ray projection data during the helical shuttle scan and the variable pitch helical scan. Further, by using the helical scan, the conventional scan, or the cine scan, it is possible to improve the accuracy and the image quality of the tomographic image reconstructed.

In step P4, the preprocessing unit 31 performs preprocessing. Here, pre-processing is performed on the X-ray detector data D0 (view, j, i) to convert it into projection data. Specifically, offset correction is performed, logarithmic conversion is performed, X-ray dose correction is performed, and sensitivity correction is performed.
In step P5, the beam hardening processing unit 33 performs beam hardening correction. Here, beam hardening correction is performed on the preprocessed projection data D1 (view, j, i). At this time, since independent beam hardening correction can be performed for each j column of the detector, if the tube voltage of each X-ray data acquisition system varies depending on the imaging conditions, the X-ray energy characteristics of the detector for each column Can be corrected.

  In step P6, the image reconstruction unit 34 performs z filter convolution processing. Here, a z-filter convolution process for applying a filter in the z-direction (column direction) is performed on the projection data D11 (view, j, i) subjected to beam hardening correction. That is, the multi-row X-ray detector D11 (view, j, i) (i = 1 to CH, j = 1 to ROW) subjected to beam hardening correction after pre-processing in each view angle and each X-ray data acquisition system For example, a filter with a column direction filter size of 5 columns is applied in the column direction.

  By controlling the column-direction filter coefficient in each of the central channel and the peripheral channel of the multi-row X-ray detector 24, the slice thickness can be controlled in each of the central portion and the peripheral portion. When the slice thickness is slightly reduced with the row direction filter, both artifacts and noise are greatly improved. Thereby, artifact improvement and noise improvement can also be controlled. That is, it is possible to control the tomographic image reconstructed, that is, the image quality in the xy plane. As another embodiment, a tomographic image having a thin slice thickness can be realized by using a deconvolution filter with column-direction (z-direction) filter coefficients. Further, the fan beam X-ray projection data is converted into parallel beam X-ray projection data as necessary.

  In step P7, the image reconstruction unit 34 performs reconstruction function superimposition processing. That is, the Fourier transform (Fourier Transform) for transforming the X-ray projection data into the frequency domain is performed, the reconstruction function is multiplied, and the inverse Fourier transform is performed.

  In step P8, the image reconstruction unit 34 performs a three-dimensional backprojection process. Here, three-dimensional backprojection processing is performed on the projection data D13 (view, j, i) subjected to reconstruction function superimposition processing to obtain backprojection data D3 (x, y, z). The image to be reconstructed is a plane perpendicular to the z axis. A three-dimensional image is reconstructed on the xy plane. The following reconstruction area P is assumed to be parallel to the xy plane. This three-dimensional backprojection process will be described later with reference to FIG.

In step P9, the image reconstruction unit 34 performs post-processing. Post-processing such as image filter superposition and CT value conversion is performed on the backprojection data D3 (x, y, z) to obtain a tomographic image D31 (x, y, z).
In step P10, the tomographic image reconstructed is displayed. As an example of the tomographic image, a tomographic image 42 is shown on the right side of FIG.
In Step P11, three-dimensional image display or MPR (Multi Plain Reformat) image display is performed. Here, a three-dimensional image 43 and an MPR image 44 in which tomographic images continuously taken in the z direction are displayed by a three-dimensional MIP (Maximum Intensity Projection) image display method are shown. Although there are various other image display methods, the operator uses different image display methods appropriately for diagnosis purposes.

<Flowchart of 3D backprojection processing>
FIG. 3 is a flowchart showing details of the three-dimensional backprojection process (step S8 in FIG. 2). In the present embodiment, the image to be reconstructed is reconstructed into a three-dimensional image on a plane perpendicular to the z axis and an xy plane. That is, the following reconstruction area is assumed to be parallel to the xy plane.

  In step P81, all views necessary for image reconstruction of tomographic images, that is, 360-degree views or 180-degree + X-ray fan beam projection data for fan angles, or fan-para-transformed X-rays. In the case of parallel beam projection data, attention is paid to all views of 360 degrees or one view of all views of 180 degrees, and projection data Dr corresponding to each pixel in the reconstruction area is extracted.

  The X-ray transmission direction is determined by the X-ray focal point of the X-ray tube 21 and the geometric position of each pixel and the multi-row X-ray detector 24. The z-coordinate z (view) of the X-ray detector data D0 (view, j, i) is known to be attached to the X-ray detector data as a table linear movement z-direction position Ztable (view). Therefore, the X-ray transmission direction is accurately determined even in the X-ray detector data D0 (view, j, i) during acceleration / deceleration in the focal point of the X-ray tube 21 and the data acquisition system of the multi-row X-ray detector 24. Can be requested.

  When a part of the line goes out of the channel direction of the multi-row X-ray detector 24, the corresponding projection data Dr (view, x, y) is set to “0”. Further, if the projection is out of the z direction, the projection data Dr (view, x, y) is extrapolated.

  In step P82, the projection data Dr (view, x, y) is multiplied by cone beam reconstruction weighted addition coefficients ωa, ωb to generate projection data D2 (view, x, y). By adding the cone beam reconstruction weighted addition coefficients ωa and ωb and adding, cone angle artifacts can be reduced. Alternatively, in the case of re-beam image reconstruction, each pixel on the reconstruction area is further multiplied by a distance coefficient. In the distance coefficient, the distance from the focus of the X-ray tube 21 to the detector row j and the channel i of the multi-row X-ray detector 24 corresponding to the projection data Dr is r0, and from the focus of the X-ray tube 21 to the projection data Dr. When the distance to the pixel on the corresponding reconstruction area P is r1, (r1 / r0) 2. In the case of parallel beam image reconstruction, each pixel on the reconstruction area P may be multiplied by only the cone beam reconstruction weighted addition coefficient w (i, j). Note that ωa + ωb = 1.

In Step P83, the projection data D2 (view, x, y) is added to the back projection data D3 (x, y) in correspondence with the pixels. Specifically, the projection data D2 (view, x, y) is added in correspondence with the pixels to the backprojection data D3 (x, y) that has been cleared in advance. The diagram shown on the right side of FIG. 3 shows the concept of adding projection data D2 (view, x, y) for each pixel.
In Step P84, it is determined whether or not the back projection data D2 of all views necessary for image reconstruction are added. Here, if not all are added, steps S81 to S81 are performed for all views necessary for image reconstruction of a tomogram (that is, views for 360 degrees or views for "180 degrees + fan angle"). S83 is repeated and all views necessary for image reconstruction are added. If all the values have been added, this process ends.

<< body movement prevention photography method >>
The imaging method of the present embodiment performs imaging that prevents body movement as described below.
FIG. 4 is a diagram showing a plurality of examples for switching the X-ray tube voltage.
In FIG. 4A (a), the first shooting time t1 and the second shooting time t2 are continuously shot. In this case, the X-ray controller 22 changes the X-ray tube voltage between time t1 and time t2. At this time, if the fan angle of the X-ray fan beam of the multi-row X-ray detector 24 is 60 degrees, the X-ray data collection angle fan angle +180 degrees necessary for the half scan H-Scan is 240 degrees, and 2/3 It becomes the amount of rotation. Now, if the rotational speed of the X-ray data acquisition system is 0.35 seconds / rotation, the imaging time is 0.46 seconds, and the body movement of the subject is considerably suppressed. Note that in the imaging method, the order of the first scan and the second scan may be reversed.

FIG. 4A (b) shows an example in which the time during which the X-ray is turned off in order to increase the X-ray tube voltage is set between the first imaging time t1 and the second imaging time t2.
In the imaging method shown in FIG. 4A (b), after the imaging time t1 of the first low X-ray tube voltage, Δt ISD (Inter Scan
During (Delay), the X-ray is turned off to prepare for changing the high X-ray tube voltage. Thereafter, the imaging time t2 of the second high X-ray tube voltage is imaged from the same X-ray data collection start angle as the first imaging. In this way, processing on the X-ray projection data can be performed with high accuracy by matching the X-ray data collection start angles.
If the rotation speed of the X-ray data acquisition system is 0.35 seconds / rotation and half scan H-Scan is performed, the imaging time is 0.58 seconds. Since the imaging time is short, the body movement of the subject is considerably suppressed.

The imaging method shown in FIG. 4B (c) changes the X-ray tube voltage for each view by collecting the odd-numbered view with the X-ray data of 140 kV and collecting the even-numbered view with the X-ray data of 80 kV. Reduce the movement of the subject.
As a modification thereof, the imaging method shown in FIG. 4B (d) alternately collects X-ray data with an X-ray tube voltage of 140 kV and X-ray data with an X-ray tube voltage of 80 kV for each of a plurality of unit views. Do.

FIG. 5 is an outline of the image reconstruction process when the X-ray tube voltage is switched for each view or for each of the plurality of views shown in FIG. 4B (c) or FIG. 4B (d). FIG. 6 is a flowchart of the image reconstruction process.
In step C1, an odd-numbered view is imaged with a low X-ray tube voltage and an even-numbered view with a high X-ray tube voltage, and X-ray data collection is performed.
In step C2, low X-ray tube voltage projection data R-Low and high X-ray tube voltage projection data R-High are separated.
In step C3, the low X-ray tube voltage projection data R-Low is reconstructed to obtain a tomographic image TG-L having a low X-ray tube voltage.
In step C4, high X-ray tube voltage projection data R-High is reconstructed to obtain a tomographic image TG-H having a high X-ray tube voltage.

In step C5, the tomographic image TG-L with a low X-ray tube voltage is multiplied by a weighted addition coefficient w1, and the tomographic image TG-H with a high X-ray tube voltage is multiplied by a weighted addition coefficient -w2. By adding them, a dual energy imaging tomographic image M-TG is reconstructed.
In step C6, a dual energy imaging tomographic image M-TG is displayed.
Note that even when X-ray data collection of each voltage is performed in units of a plurality of views, the image may be reconstructed by separating into R-Low and R-High.

  As described above, the imaging method for switching the X-ray tube voltage at an early timing can obtain a dual energy tomographic image with little body movement.

<Operation of the tomographic image position correction unit 37>
Hereinafter, the operation of the tomographic image position correction unit 37 will be described in detail using an embodiment.
(Example 1) <Position displacement correction using sector area FA>
Next, an example of performing alignment between the tomographic image TG-L photographed with a low X-ray tube voltage and the tomographic image TG-H photographed with a high X-ray tube voltage using a sector area FA determined by three points. Will be described.
In general, the human body is composed of hard objects such as bones and soft objects such as soft tissues and internal organs. Since the skeleton does not move much with respiration and body movement, soft regions such as soft tissues and internal organs are expected to move around this hard region. Also, the bone region is easy to extract because it has a large X-ray absorption coefficient and a high CT value. For this reason, the alignment is performed by extracting a bone region having a large X-ray absorption coefficient of the CT value and using the center or the center of gravity as the center.
At least three points are required for alignment in a two-dimensional plane. For this reason, it is divided into triangular or fan-shaped areas defined by three points, and alignment is performed. At this time, one of the three points is set to the center of the bone region that is the reference point that does not move. Positioning is performed using the three fan-shaped areas FA.

  FIG. 7 is a flowchart for performing alignment in a sector (radial) region when reconstructing a two-dimensional distribution tomographic image of X-ray tube voltage-dependent information in the X-ray absorption coefficient. FIG. 8 is a diagram illustrating extraction of a bone region. FIG. 9A is a diagram in which a part of the sector area FA is overlapped, and FIG. 9B is a diagram illustrating an addition coefficient in each of the overlapping sector areas FA.

In 8 steps D1 shown in FIG. 7, the X-ray data collection unit performs imaging at an X-ray tube voltage of 80 kV.
In step D2, the X-ray data collection unit performs imaging with an X-ray tube voltage of 140 kV.
In step D3, in each of the 80 kV and 140 kV tomographic images, the tomographic image position shift correction unit 37 performs binarization processing on each tomographic image that is a grayscale image with a threshold value for extracting bones. As shown in FIG. 8, each tomographic image of a grayscale image is assumed to be, for example, 150 HU (Hounsfield expected as a CT value of bone
The binarization process is performed using the region above (Unit) as a threshold.

In step D4, the tomographic image position correction unit 37 performs a continuous area numbering process (labeling process). As shown in FIG. 8, a labeling process is performed on the binarized image, and numbering is performed for each two-dimensional region composed of continuous pixels. As shown in the enlarged view of FIG. 8, the pixel value of each continuous two-dimensional region is a continuous region number value, and here, each pixel is “2”.
In Step D5, the tomographic image position correction unit 37 obtains an image feature amount for each continuous region. That is, for each continuous two-dimensional region numbered in step D4, an image such as an average CT value, a standard deviation of CT values, an area value, a circularity, a ferret diameter, an area ratio, a long diameter, a short diameter, or flatness Find the features. These image feature amounts will be described later with reference to FIG.

In step D6, the tomographic image position correction unit 37 extracts a bone region from each continuous region. Here, as shown in FIG. 8, a region that is not a bone region, for example, a contrast agent region, is removed based on the image feature amount obtained for each continuous two-dimensional region in step D5.
In step D7, the tomographic image position correction unit 37 determines a bone part region tn as a center based on the feature parameters of each bone part region. Normally, in the neck, chest, abdomen, and lumbar region, select the bone region centered on the vertebral body, the femur and tibia in the lower limb (foot), the humerus and rib in the upper limb (hand), and the head It is recommended to select the bone region centered on the skull. Further, the center of gravity o of this bone region tn is obtained.

  In step D8, the tomographic image position correction unit 37 performs binarization processing with a threshold value that can extract the contours of the subject body surface of both tomographic images. Binarization processing is performed with a threshold value that can extract the contour of the body surface from tomographic images with X-ray tube voltages of 80 kV and 140 kV, for example, a threshold value of about −100 HU to −200 HU.

In step D9, the tomographic image position correction unit 37 extracts the contour of the body surface, divides the contour of the body surface into 8 parts with radiation every 45 degrees centered on the center of gravity o, and is divided into eight divided points a to h. Determine. The contour line of the body surface is obtained by, for example, a contour extraction logic filter. As shown in FIG. 8, the center of gravity o (gx, gy) of the bone region tn is the center, and 0 degrees, 45 degrees, 90 degrees, 135 degrees, 180 degrees, 225 degrees, 270 degrees, Intersections between the dividing lines of 315 degrees every 45 degrees and the contour lines of the body surface are obtained, and these are set as points a, b, c, d, e, f, g, h.
Here, the points a to h of the tomographic image TG-L with the X-ray tube voltage of 80 kV are designated as points a80 to h80, and the points a to h of the tomographic image TG-H with the X-ray tube voltage of 140 kV are designated as points a140 to It is set as h140. In this embodiment, the dividing lines are set every 45 degrees. However, when the positional deviation between X-ray tube voltages of 80 kV and 140 kV is large between the tomographic images, the dividing lines every 45 degrees are set to 30 degrees. You may divide | segment and set more finely like 22.5 degree | times and 10 degree | times.

In step D10, the similarity ratio between Δg80h80o and Δg140h140o is obtained. On the tomographic image TG-L having a tube voltage of 80 kV, Δa80b80o to Δh80a80o are defined, and on the tomographic image TG-H having a tube voltage of 140 kV, eight fan-shaped regions FA are defined. In this case, the line segments constituting each fan-shaped area FA of both tube voltages are always in the same direction, and in this case, only the scaling factor should be considered. Then, as shown in FIG. 8, the ratio of the sides is obtained in each sector area FA. That is, the ratio rg between the line segment g80-o and the line segment g140-o and the ratio rh between the line segment h80-o and the line segment h140-o are obtained.
In step D11, the tomographic image position shift correction unit 37 performs enlargement / reduction coordinate conversion of Δg80h80o with a similar shape ratio to match Δg140h140o. When the fan-shaped linear portions at every 45 degrees are used as axes, the respective fan-shaped positions can be aligned by performing the enlargement / reduction ratio processing for the fan-shaped linear portions at every 45 degrees in the axial direction by rg and rh.

Note that the coordinate transformation by affine transformation is performed with respect to the enlargement / reduction magnification in each 45 degree direction obtained in step D11 as follows. For example, consider the direction of oa and the direction of ob.
When the ratio ra between the line segment a80-o and the line segment a140-o and the ratio rb between the line segment b80-o and the line segment b140-o are obtained, conversion is performed by the affine transformation of the following (Formula 1). .
However, the coordinates before conversion are (x, y), and the coordinates after conversion are (X, Y).
... (Formula 1)
Further, this process can be realized not only by affine transformation but also by multi-dimensional coordinate transformation. When multi-order coordinate transformation is performed, coordinate transformation with higher accuracy than affine transformation, which is primary coordinate transformation, can be performed.
The coordinate conversion performed in step D11 may be performed so that the sector area FA with an X-ray tube voltage of 80 kV matches the sector area FA with an X-ray tube voltage of 140 kV, or vice versa.

In Step D12, the tomographic image position correction unit 37 determines whether or not coordinate transformation alignment has been performed for Δaho, Δabo, Δbco, Δcdo, Δdeo, Δefo, Δfgo, and Δgho. If YES, the process goes to Step D12. If NO, the process returns to step D9.
In step D13, the tomographic image position correction unit 37 combines the sector areas FA of the 80 kV tomographic image TG-L and the 140 kV tomographic image TG-H. The fan-shaped areas FA with the X-ray tube voltages of 80 kV and 140 kV are combined. At this time, there may be a case where the tomographic images of the joint portions that combine the fan-shaped areas FA are not continuously joined. In order to avoid this, processing is performed in an area larger than each sector area FA. For example, as shown in FIG. 9A, the process is performed by overlapping the sector area FAA and the sector area FAB. At the boundary between the fan-shaped areas FAA and B, the tomographic images obtained by aligning the X-ray tube voltages are added to the addition coefficients wa (x, y) and wb (x, y) as shown in FIG. To process. The sum of the coefficients wa (x, y) and wb (x, y) is always “1” and constant.
By performing the weighted addition process on the tomographic images from the sector area FAA to the sector area FAH in which the weighted addition coefficients are aligned, the images are continuously and smoothly joined. At this time, a linear weighted addition coefficient is used, but this may be a multi-degree polynomial weighted addition coefficient.

In step D14, the dual energy image reconstruction unit combines the fan-shaped areas FA of both X-ray tube voltages, and performs weighted addition processing of the 80 kV tomographic image TG-L and the 140 kV tomographic image TG-H.
In Step D15, the dual energy imaging tomographic image M-TG obtained in Step D14 is displayed.
In the processing from step D3 to step D7, by obtaining the center of gravity o of the bone area of each tomographic image, the “fan key” of each sector area FA determined after step D8 is determined and can be divided into sectors. We are preparing.

<Example of image feature amount>
FIG. 10 shows an example of the image feature amount used in step D5.
First, assuming that a continuous two-dimensional region is Rn, an area s of a pixel unit of the region Rn and an actual area sp (unit mm ² ) are expressed by the following (Equation 2) and (Equation 3).
... (Formula 2)
... (Formula 3)
The average CT value m and its standard deviation sd are as shown in the following (Equation 4) and (Equation 5). However, g (x, y) is a pixel value of coordinates (x, y).

... (Formula 4)
... (Formula 5)

Further, the circularity c is expressed by the following (Equation 6). Here, l is the peripheral length of the region Rn.
... (Formula 6)
Further, the ferret diameter is lx in the x-direction fillet diameter and ly in the y-direction from the circumscribing of the region Rn, and the area ratio sr is expressed by (Equation 7).
... (Formula 7)
Further, the flattening ratio R is obtained as in (Equation 8) from the major axis Rx and the minor axis Ry that are elliptically approximated in the region Rn.
... (Formula 8)

(Embodiment 2) <Position displacement correction 1 using profile data of line segments extending radially from the center of gravity>
In Example 1, the alignment of the fan-shaped area FA of the tomographic image regarding both tube voltages was performed at three vertices defined by the intersections of the contour line of the subject body surface and the dividing line. In contrast, the present embodiment is a method of matching pixels on a line segment extending radially from the center of gravity with a pixel value, that is, a method using a profile curve.
FIG. 11A is a diagram showing profile data on two sides of each sector area FA. By aligning the profiles prf80 and prf140 in the form of profiles, a more accurate weighted addition process can be realized.

First, the tomographic image position correction unit 37 obtains the shift amount s1 and the enlargement / reduction ratio r1 by the following method while matching the corresponding pixels of the profiles prf80 and prf140.
In general, an air portion in an X-ray CT tomogram takes a CT value of −1000, and a subject portion whose X-ray absorption coefficient is substantially equivalent to water takes a CT value of 0 or around 0.
The profiles prf80 and prf140 representing changes in the CT value are calculated as profiles in which 1000 is added to the CT value of each pixel so that minus 1000 of the air portion does not become minus and the minus (minus) portion is eliminated. Alternatively, the negative part of the CT value may be calculated by compressing the CT value to zero. By doing so, it is not necessary to include a calculation error due to the negative part. This correlation calculation is as shown in (Formula 9) below.
... (Formula 9)

The tomographic image position correction unit 37 obtains the result Ca (s) of the correlation calculation between the profile prf80 and the profile prf140 on the line segments a80o and a140o.
FIG. 11B shows the correlation of profile data. A distance s1 in FIG. 11B indicates the amount of deviation at the peak of Ca (s). Further, the enlargement / reduction ratio r1 of the profile prf140 with respect to the profile prf80 is determined by the following equation (Equation 10). s1 indicates a deviation amount between the profile prf80 and the profile prf140.
... (Formula 10)
Here, s2 is a half width p0 / 2 of the peak value p0 of the correlation calculation Ca (s), that is, a full width half maximum (FWHM), and l80 = a80o. Using this positional deviation amount s1 or the enlargement / reduction ratio r1, coordinate conversion similar to that in the first embodiment can be performed.

FIG. 12 is a flowchart showing the process of combining the fan-shaped areas FA of the tube voltages using the correlation calculation.
After step D9 in FIG. 7 is completed, the process proceeds to step D25.
In step D25, a profile correlation Ca (s) on two line segments in each sector area FA is obtained.
In step D26, the peak position shift amount s1 or the enlargement / reduction ratio r1 is obtained from the correlation calculation result in step D25.
In step D27, each sector area FA of 80 kV and each sector area FA of 140 kV are aligned by coordinate conversion.
In step D28, it is determined whether or not alignment has been completed for all the fan-shaped areas FA. If YES, the process proceeds to step D13 in FIG. 7, and if NO, the process returns to step D25.
In the above embodiment, the correlation calculation is performed for two sides of each sector area FA. However, when more correlation calculations are desired, the line segment may be increased in the angular direction as shown in FIG. . Further, by obtaining the peak shift amount s1 or the enlargement / reduction ratio r1 on the line segment based on the correlation calculation result of the increased line segment, coordinate conversion with higher accuracy can be performed.

(Example 3) <Position displacement correction 2 using profile data of line segments extending radially from the center of gravity>
In Example 2, the peak position deviation amount s1 or the enlargement / reduction ratio r1 was obtained by obtaining the profile data correlation Ca (s) on the two sides of the sector area FA having different tube voltages. However, only a part of the body of the subject may be enlarged or reduced. The linear deviation correction described here is a method of eliminating a state in which a certain part of the profile is extended.
FIG. 13A shows the profile prf80 and profile prf140 of each sector area FA, and FIG. 13B shows the profile data of 80 kV and the profile data of 140 kV. (D) It is a figure which shows the alignment of Embodiment 2. FIG. (Figure explanation P57)
FIG. 14A is a diagram showing the amount of positional deviation of the peak in the diameter direction of the profile of each sector area FA. (B) is a figure which shows the relationship between the integral value of the positional offset amount of a peak, and a diameter direction coordinate.

The region r1 in the y direction of the profile prf80 has a strong correlation with the region r1 in the y direction of the profile prf140. The region r2 in the y direction of the profile prf80 has a strong correlation with the region r2 in the y direction of the profile prf140. However, if correlation calculation processing is performed on the entire line segment h80o and line segment h140o, the strong correlation between the regions r1 and r2 is correlated due to the profile data of the partially extended r3 region. Shifts. This appears as a misalignment in the diameter direction, which is a direction extending radially from the center of gravity, in the alignment of each sector area FA.
In this way, in order to extract a portion that shows a partial strong correlation in the radial profile of each sector area FA, each sector area FA may be further divided in the diameter direction.

For each of the radial regions of the sector regions h80a80o and the sector regions h140a140o shown in FIG. 13A, the tomographic image positional deviation correction unit 37 correlates the profile in the diametrical direction, and the peak positional deviation amount s1, or enlargement / reduction A magnification r1 is obtained.
First, with profile data representing a change in CT value outward in the radial direction in the radial direction r, the profile of the tomogram TG-L with an X-ray tube voltage of 80 kV is taken as prf80 (r), and the tomogram with an X-ray tube voltage of 140 kV. When the profile of TG-H is prf140 (r), a partial correlation function Co (x) is obtained by the following (Formula 11).
... (Formula 11)

However, Δr is a value determined from the frequency and period of profile change, and in the case of a subject, Δr may be about 2 to 5 mm.
At this time, the correlation function considers how much the position of the diametrical coordinate r is shifted to reach the maximum value (peak), that is, whether the profile prf80 (r) and the profile prf140 (r) are highly matched.
In the diameter direction coordinate position r1 in FIG. 14A, the correlation function C1 (r, s) of the following (Equation 12) takes the maximum value with the positional deviation amount s1.
... (Formula 12)
A positional deviation amount s that takes the peak of the correlation function at each diametrical coordinate r, that is, a positional deviation amount s that closely matches prf80 (r) and prf140 (r) is obtained. As shown in FIG. 14B, the shift amount s that takes a peak is a function of the diameter direction coordinate r.

FIG. 15 is a diagram showing the shift amount s as shift (r). FIG. 15A shows the profile prf80 (r), and FIG. 15B shows the profile prf140 (r). FIG. 15C shows the integrated value of the positional deviation amount s, FIG. 15D shows the local deviation amount Δs, and FIG. 15E shows the sector area FA divided in the diameter direction. FIG.
The shift (r) shown in FIG. 15C is an integral value of the shift amount s from the line segment a80o, the line segment a140o, the line segment h80o, or the line segment h140o to the diameter direction coordinate r. Further, the local deviation amount Δs in [r, r + Δr] shown in FIG. 15D is a derivative of the integral value shift (r) of this deviation. The following (Equation 13) is the local deviation amount Δs.
... (Formula 13)
The tomographic image position shift correction unit 37 divides the region in the r direction for each range where the shift amount Δs is somewhat the same. In the example of FIG. 15D, the area is divided into r-direction areas r1 to r6. When this is developed in the sector area FA, it becomes as shown in FIG. For each region from the region F1 to the region F6, the sector-shaped region FAhao is aligned by performing alignment by shifting the amount of deviation Δs in the diameter direction by coordinate conversion. Further, discontinuous artifacts on each boundary line of each region can be prevented by overlapping the regions as in the first embodiment.

FIG. 16 is a flowchart of the correction of displacement in the diameter direction with the above profile curve.
After step D9 in FIG. 7 is completed, the process proceeds to step D45.
In step D45, the tomographic image position deviation correction unit 37 obtains a deviation amount of each position in the diametrical direction using a profile on two lines or a plurality of diametrical line segments in each sector area FA.
In step D46, each sector area FA is divided in the diameter direction according to the amount of deviation.
In step D47, the region divided in the diameter direction in each sector region FA of each X-ray tube voltage is aligned by coordinate conversion.
In step D48, it is determined whether or not the alignment is completed in all sector areas FA. If YES, the process proceeds to step D13 in FIG. 7, and if NO, the process returns to step D45.

  Through the above processing, the tomographic image position correction unit 37 extracts the position of a part having a strong correlation in the diameter direction profile data in each sector area FA of both X-ray tube voltages. Then, due to the difference in the amount of deviation Δs, alignment can be performed with higher accuracy by dividing the sector area FA in the diameter direction and performing alignment. Specifically, alignment of a positional deviation that is not uniform in the radial direction of the body axis, that is, the diameter direction of the fan-shaped area FA, can be realized with high accuracy by the respiration, heartbeat, pulsation, and the like of the subject.

(Example 4) <Position shift correction using two-dimensional correlation calculation>
In the second embodiment, the positional deviation amount and the enlargement / reduction ratio are determined only by the three vertices of the sector area FA. In addition, the correlation calculation of the profile data of the radial line segment extending outward from the center in the sector area FA is also a one-dimensional correlation calculation.
In the present embodiment, the tomographic image position correction unit 37 divides a tomographic image of one X-ray tube voltage into a sector area FA and performs a two-dimensional correlation operation on the tomographic image of the other X-ray tube voltage. Then, the tomographic image position correction unit 37 obtains the x-direction shift amount Δx and the y-direction shift amount Δy from the position of the maximum value or the local maximum value of the two-dimensional correlation calculation, and the maximum value of the two-dimensional correlation calculation, Alternatively, the enlargement / reduction ratio is obtained from the half width of the peak of the local maximum value.

FIG. 17A is a diagram showing an outline of alignment processing by two-dimensional correlation calculation.
In the tomographic image TG-H having an X-ray tube voltage of 140 kV, the tomographic image position shift correction unit 37 detects a bone region that is supposed not to move, and a sectoral region FA at a certain angle with the center of gravity o as the center. Determine. Each pixel is g140 (x, y), and each pixel having an X-ray tube voltage of 80 kV is g80 (x, y). When the two-dimensional correlation calculation is performed on the tomogram TG-L having the X-ray tube voltage of 140 kV on the tomographic image TG-L with the X-ray tube voltage of 80 kV, the correlation amount in the fan-shaped region FAh140a140o is expressed by the following (Formula 14). However, the sector area FAh140a140o is Fa140, and the sector area FAh80a80o is Fa80.
... (Formula 14)
FIG. 17B shows an example of the change in the correlation amount. The correlation amount takes a peak at a position shifted by (Δx, Δy) from the origin of the xy plane, and its half-value width is d3x in the x direction and d3y in the y direction.

Δx and Δy indicate a deviation between the sector area FAh140a140o and the sector area FAh80a80o. Further, from the half width, the enlargement / reduction ratios r3x and r3y in the x direction and the y direction are determined from the following (Expression 15) and (Expression 16). In the following formula, l140x and l140y are the lengths of the sides in the x and y directions circumscribing the sector area FAh140a140o.
... (Formula 15)
... (Formula 16)
From the enlargement / reduction magnifications r3x, r3y and the shift amounts Δx, Δy, the sector area FA having the X-ray tube voltage 140 kV is aligned with the tomographic image TG-L having the X-ray tube voltage 80 kV.
By performing the above processing from the sector areas FAa140b140o to the sector areas FAg140h140o, the tomographic image position correction unit 37 can align all the sector areas FA with the tomogram TG-L having an X-ray tube voltage of 80 kV.

FIG. 18 is a flowchart of misalignment correction by two-dimensional correlation calculation.
After step D9 in FIG. 7 is completed, the process proceeds to step D65.
In step D65, the tomographic image position correction unit 37 performs a two-dimensional correlation calculation on the tomographic image TG-L with the X-ray tube voltage of 80 kV for the sector area FA with the X-ray tube voltage of 140 kV.
In step D66, the tomographic image position correction unit 37 obtains the position shift amount and the enlargement / reduction ratio from the position of the local maximum value of the two-dimensional correlation calculation.
In Step D67, the tomographic image position correction unit 37 determines whether or not the alignment of all the sector areas FA of the tomographic image TG-H having the X-ray tube voltage of 140 kV is completed. If YES, the process proceeds to step D13 in FIG. 7, and if NO, the process returns to step D65.
In this way, the tomographic image position correction unit 37 divides the tomographic image TG-H with the X-ray tube voltage 140 kV into the sector area FA, and performs a two-dimensional correlation calculation on the tomographic image TG-L with the X-ray tube voltage 80 kV. I do. Then, the direction deviation amounts Δx and Δy are obtained from the position of the maximum value, and the enlargement / reduction magnification is obtained from the half width. Then, coordinate conversion is performed, and the tomographic image TG-H having an X-ray tube voltage of 140 kV and the tomographic image TG-L having an X-ray tube voltage of 80 kV are subjected to weighted addition processing, so that the tomographic image M- of dual energy imaging is obtained. TG is obtained.
In the present embodiment, the tomographic image TG-H with the X-ray tube voltage of 140 kV is divided into the sector areas FA, and the two-dimensional correlation calculation is performed with the tomographic image TG-L with the X-ray tube voltage of 80 kV. A tomographic image TG-L having a voltage of 80 kV may be divided into sector regions FA and a two-dimensional correlation operation may be performed with the tomographic image TG-H having an X-ray tube voltage of 140 kV. As described above, even when performing a two-dimensional correlation operation on the fan-shaped region in the tomographic image of one X-ray tube voltage and the tomographic image of the other X-ray tube voltage, the “first tomographic image and It corresponds to the fact that the center of gravity common to the second tomographic image is obtained and compared with a comparison target set in a direction extending radially from the center of gravity.

(Example 5) <Manual displacement correction 1>
Considering the following cases, in the present embodiment, the alignment is performed manually by visual observation.
(1) When periodic artifacts such as streak noise exist in the tomographic image and an error in alignment is likely to occur due to correlation calculation processing. (2) Image quality of the tomographic image is poor, and alignment due to noise or the like causes alignment. If the error is likely to occur

FIG. 19 shows an outline of manual alignment operation.
First, in the tomographic image TG-H having an X-ray tube voltage of 140 kV, the operator manually sets the center (center of gravity) of the central bone region, and the direction of the dividing line extending radially from the center Also set manually.
After that, the operator manually defines each sector area FA with the two outer points. The fan-shaped area FA determined manually is displayed on the monitor 6 in an overlapping manner on the tomographic image TG-L having an X-ray tube voltage of 80 kV. By moving the operator by dragging and dropping or the like, the operator moves the two points outside the fan-shaped area FA and the fan-shaped area FA of the tomographic image TG-H having the X-ray tube voltage of 140 kV to the tomographic image having the X-ray tube voltage of 80 kV. The alignment can be performed so as to overlap the image TG-L.
This operation is performed on all the sector areas FA, and the sector areas FA are combined to reproduce the tomographic image TG-H having an X-ray tube voltage of 140 kV. Thereby, it can overlap and can position.

FIG. 20 shows a flowchart of manual deviation correction.
In step D71, imaging is performed with an X-ray tube voltage of 80 kV.
In step D72, imaging is performed with an X-ray tube voltage of 140 kV.
In step D73, the center position (center of gravity) of the bone region that is considered not to move in the tomographic image TG-H having the X-ray tube voltage of 140 kV is manually determined. The “fan essential” of each fan-shaped area FA is placed in the bone area. For example, in the abdomen, since the vertebral body supports the weight of the subject, it is considered that the part does not move even when breathing or pulsating. For this reason, it is possible to select “the point of the fan” as a point where the central part of the vertebral body does not move. However, it is not always necessary to set the center in the vertebral body or the bone region, and if there is a place that seems not to move in other places, it is possible to select it as a “fan key”.

In step D74, in the tomographic image TG-H having an X-ray tube voltage of 140 kV, the direction of the line extending in the radial direction from the center position of the bone region is manually set, and each sector region FA is also set manually.
Further, in step D74, in FIG. 19, the two end points of the sector area FA arc are placed on the contour line of the subject body surface, but need not necessarily be on the contour line of the subject body surface. Further, two end points of a sector arc may be placed in the air outside the surface of the subject body.

  In step D75, the tomographic image position correction unit 37 overlies the sector area FA of the tomographic image TG-H having the X-ray tube voltage of 140 kV on the tomographic image TG-L having the X-ray tube voltage of 80 kV displayed on the monitor 6. Display a lap. Therefore, the operator manually aligns the sector area FA. When aligning the fan-shaped area FA of the tomographic image TG-H having an X-ray tube voltage of 140 kV displayed on the monitor 6 with the tomographic image TG-L having an X-ray tube voltage of 80 kV, The three vertices are moved on the monitor 6 by a drag and drop operation or the like, and the degree of overlap after the movement is visually determined. Further, the operator moves the sector area FA to obtain a point where the sector area FA with the X-ray tube voltage of 140 kV and the tomographic image TG-L with the X-ray tube voltage of 80 kV overlap well. The tomographic image position correction unit 37 determines from which position (xsi, ysi) each of the three vertexes of the sector-shaped area FA of the tomographic image TG-H having an X-ray tube voltage of 140 kV (where i = 1, 2, 3). The position (xei, yei) that has been moved is stored.

In step D76, it is determined whether or not the alignment of all the sector areas FA of the tomographic image TG-H having the X-ray tube voltage of 140 kV is completed. If YES, the process goes to step D77, and if NO, the process returns to step D75.
In step D77, the sector areas FA of the tomographic image TG-H having the X-ray tube voltage of 140 kV are merged. Each sector area FA is overlapped as shown in FIG. 9A, and the weighted addition coefficient is continuously changed in the overlapped area as shown in FIG. 9B. By doing so, it is possible to make the tomographic image by the combined fan-shaped areas FA in Step D77 a continuous tomographic image without discontinuous artifacts.
In step D78, a weighted addition process is performed on the tomographic image TG-L having an X-ray tube voltage of 80 kV and the tomographic image TG-H having an X-ray tube voltage of 140 kV.
In step D79, a tomographic image M-TG for dual energy imaging is displayed.

(Example 6) <Manual deviation correction 2>
FIG. 21 shows an outline of an alignment operation that further simplifies the operation without using a dividing line extending in the radial direction.
In the tomographic image TG-H having an X-ray tube voltage of 140 kV, the center o (center of gravity) of the bone region is set manually. Next, a plurality of points are manually set on the body surface outline of the subject.
The tomographic image position shift correction unit 37 displays the tomographic image TG-H with the X-ray tube voltage 140 kV and the tomographic image TG-H with the X-ray tube voltage 80 kV on the monitor 6 in an overlapping manner. The operator then aligns the body surface contour point of the X-ray tube voltage 140 kV tomographic image or a plurality of points close to the body surface contour line on the body surface contour line of the X-ray tube voltage 80 kV tomographic image. This makes it possible to align the tomographic images of both X-ray tube voltages. In the above, an embodiment in which the tomographic image TG-H having an X-ray tube voltage of 140 kV is combined with the tomographic image TG-H having an X-ray tube voltage of 140 kV is described. It may be adjusted to the 80 kV tomographic image TG-L.

FIG. 22 is a flowchart of alignment dual energy imaging processing that does not use manual dividing lines.
After step D73 in FIG. 20 is completed, the process proceeds to step D84.
In step D84, a plurality of points are manually set in the tomographic image TG-H having an X-ray tube voltage of 140 kV on the contour line of the body surface of the subject or in the vicinity thereof. In the case of FIG. 19, a plurality of points are placed on the contour line of the subject body surface, but it is not always necessary to place a plurality of points on the contour line of the subject body surface, and a plurality of points are placed in the outside air. May be.

In step D85, the tomographic image position correction unit 37 places the tomographic image TG-H having the X-ray tube voltage 140 kV and the body surface of the subject on the tomographic image TG-L having the X-ray tube voltage 80 kV displayed on the monitor 6. A plurality of points set at or near the contour line of is displayed. Then, the operator manually adjusts the plurality of points. A plurality of points set on or near the contour line of the surface of the subject are moved on the monitor 6 by a drag and drop operation or the like, and a tomographic image TG-H having an X-ray tube voltage of 140 kV and an X-ray tube voltage of 80 kV. The point which makes the tomographic images TG-L of the two overlap well is obtained. In this case, as shown in FIG. 22 (b), there are a center point o of the bone region and points h, a, and b placed on the body surface contour line of the subject. When a is moved to the point a ′, Δhao and Δabo on the tomographic image TG-H with the X-ray tube voltage of 140 kV change to Δhao and Δabo. At this time, the pixels included in Δhao and Δabo on the tomographic image are also coordinate-transformed and moved to Δhao and Δabo.
In step D86, it is determined whether or not the tomographic image TG-H having an X-ray tube voltage of 140 kV is completely aligned with the tomographic image TG-L having an X-ray tube voltage of 80 kV. If YES, the process goes to step D78 in FIG. If there is, return to Step D85.

<S / N ratio of tomographic image M-TG of dual energy imaging>
In this embodiment, when obtaining a dual energy imaging tomogram from a plurality of tomograms of X-ray tube voltages, one of the weighted addition coefficients of the weighted addition process is a negative number (minus). For this reason, compared with the original tomographic images of a plurality of X-ray tube voltages, the SN of the dual energy imaging tomographic image is deteriorated, the image noise is deteriorated, or the image quality is deteriorated. For this reason, the imaging conditions of the original tomographic images of a plurality of X-ray tube voltages must be determined in consideration of the exposure of the subject and also the image noise of the tomographic image of dual energy imaging.

FIG. 23 is a diagram illustrating image noise of the difference image. Generally, as shown in FIG. 23, a low X-ray tube voltage tomogram TG-L having an image noise of n1, a signal level of s1, and an S / N ratio of n1 / s1, an image noise of n2, and a signal level of When the difference image of the high X-ray tube voltage tomogram TG-H having the s2 and S / N ratio of n2 / s2 is obtained, the S / N ratio Nsub of the difference image is as shown in the following (Equation 17).
... (Formula 17)
From the arithmetic geometric mean theorem, the following (Equation 18) holds.
... (Formula 18)
That is, when the image noise N1 of the tomographic image TG-L having the low X-ray tube voltage kV1 and the image noise N2 of the tomographic image TG-H having the high X-ray tube voltage kV2 are equal, the image noise NSub of the difference image is minimized. .
In the case of the present embodiment, since the weighted addition coefficients w1 and w2 are included in the weighted addition process, the following equation (Equation 20) is obtained in consideration of the above.
... (Formula 20)

That is, in consideration of the weighted addition coefficient, the image noise of the tomographic image CSI-Low with the X-ray tube voltage of 80 kV and the image noise of the tomographic image CSI-High with the X-ray tube voltage of 140 kV may be substantially equal.
In addition, in so-called dual energy tomography, as a method of determining the X-ray tube voltage to obtain a better S / N ratio with as little X-ray exposure as possible, it is necessary to determine it depending on the substance to be extracted and the substance to be emphasized. There is.

  In this embodiment, an X-ray tube voltage of 80 kV and an X-ray tube voltage of 140 kV are used. However, dual energy imaging can be performed using other X-ray tube voltages. In addition, the image reconstruction method in the present embodiment may be a conventionally known three-dimensional image reconstruction method using the Feldkamp method, another three-dimensional image reconstruction method, or a two-dimensional image reconstruction.

In the present embodiment, no particular scanning method is specified. The same effect can be obtained in the case of conventional scan, helical scan, variable pitch helical scan, and helical shuttle scan. Although the present embodiment describes the case where the scanning gantry 20 is not tilted, the same effect can be obtained even in the case of so-called tilt scanning in which the scanning gantry 20 is tilted. In particular, the same effect can be obtained by synchronizing with a heartbeat signal.
In this embodiment, the medical X-ray CT apparatus is described as the original, but an industrial X-ray CT apparatus, or an X-ray CT-PET apparatus, an X-ray CT-SPECT apparatus combined with another apparatus, etc. Can also be used.

1 is a block diagram showing an X-ray CT apparatus 100 according to an embodiment of the present invention. It is a flowchart which shows the outline | summary of operation | movement about the X-ray CT apparatus of this embodiment. It is a flowchart which shows the detail of a three-dimensional backprojection process. (A) is a figure which switches X-ray tube voltage by a continuous scan, (b) is a figure which switches X-ray tube voltage by the continuous scan which has an interval. (C) is a diagram for switching the X-ray tube voltage for each view, and (d) is a diagram for switching the X-ray tube voltage for each data acquisition segment. It is a figure which shows the outline | summary of the image reconstruction at the time of changing X-ray tube voltage by odd-numbered / even-numbered view. It is a flowchart of the image reconstruction when the X-ray tube voltage is changed in the odd / even view. It is a flowchart which performs alignment in a fan-shaped (radial) area | region when image-reconstructing the two-dimensional distribution tomogram of the X-ray tube voltage dependence information in an X-ray absorption coefficient. It is a figure which shows alignment of each fan-shaped area | region FA which extracted the bone | frame part area | region and divided | segmented. (A) is the figure which overlapped a part of sector area FA, (b) is a figure which shows the addition coefficient in overlap each sector area FA. It is a figure which shows an image feature-value. (A) is a figure which showed a part of sector-shaped area | region FA, (b) is a figure which shows the correlation of profile data, (c) shows the case where the line segment which takes a correlation in an angle direction is increased. FIG. It is the flowchart which showed the process of uniting each sector area FA using correlation calculation. (A) is the figure which showed a part of sector-shaped area | region FA, (b) is a figure which shows the profile data of ay direction. (A) is a figure which shows the peak of deviation | shift amount s in the diameter direction of the profile of each sector-shaped area | region FA, (b) is a figure which shows the relationship between the integrated value of deviation | shift amount s, and a diameter direction coordinate. (A) shows the profile prf80 (r), and (b) shows the profile prf140 (r). Further, (c) shows an integrated value of the deviation amount s, (d) shows a local deviation amount Δs, and (e) shows a sector area FA divided in the diameter direction. It is a flowchart which calculates | requires the deviation | shift amount of the diameter direction of each sector area FA. (A) is a figure which shows the outline | summary of the process of the alignment by a two-dimensional correlation calculation, (b) is the result of a two-dimensional correlation calculation. It is the flowchart of the dual energy imaging | photography process which calculated | required the positional offset amount and the expansion / contraction magnification by the two-dimensional correlation calculation of each sector area FA. It is a figure which shows the outline | summary of manual alignment operation. It is a flowchart of the dual energy imaging | photography process which set each sector area FA manually and aligned manually. It is a figure which shows the outline | summary of the alignment operation which does not use a manual dividing line. (A) is a flowchart of the dual energy imaging process of the alignment which does not use a manual dividing line. (B) is a figure which shows the movement of the point when not using a dividing line. It is a figure which shows the image noise of a difference image.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Operation console 2 ... Input device 3 ... Central processing unit 5 ... Data collection buffer 6 ... Monitor 7 ... Storage device 12 ... Cradle 15 ... Rotating part 20 ... Scanning gantry 21 ... X-ray tube 22 ... X-ray controller 23 ... Collimator 24 ... Multi-row X-ray detector or two-dimensional X-ray area detector 25 ... Data acquisition device (DAS)
26: Rotating unit controller 28 ... Beam forming X-ray filter 29 ... Control controller 30 ... Slip ring 33 ... Beam hardening processing unit 34 ... Image reconstruction unit 35 ... Dual energy image reconstruction unit 37 ... Tomographic image position deviation correction unit D ... Projection data TG ... Tomographic image M-TG ... Dual energy image FA ... Fan-shaped area

Claims (13)

An X-ray tube that irradiates a subject with X-rays having a first energy spectrum and X-rays having a second energy spectrum different from the first energy spectrum;
An X-ray data collection unit for collecting the X-ray projection data of the first energy spectrum and the X-ray projection data of the second energy spectrum irradiated on the subject;
An image reconstruction unit configured to reconstruct a first tomographic image and a second tomographic image based on the X-ray projection data of the first energy spectrum and the X-ray projection data of the second energy spectrum;
The position of the first tomographic image and the second tomographic image is obtained by obtaining a centroid common to the first tomographic image and the second tomographic image and comparing with a comparison target set in a direction extending radially from the centroid. A tomographic image position correction unit that calculates a shift amount, and performs a positional shift correction based on the positional shift amount on at least one of the first tomographic image and the second tomographic image;
Dual energy image reconstruction means for reconstructing a tomographic image of X-ray tube voltage-dependent information related to the distribution of atoms based on the aligned first tomographic image and the second tomographic image. X-ray CT apparatus characterized by this. The X-ray CT apparatus according to claim 1, wherein the tomographic image position correction unit selects the center of gravity from a region having a high CT value of the subject.   The X-ray CT apparatus according to claim 2, wherein the region having a high CT value of the subject is a bone region.   The tomographic image position correction unit identifies, in each of the first tomographic image and the second tomographic image, a region obtained by dividing the tomographic image into a plurality of sectors extending radially from the center of gravity, and the first tomographic image 4. A positional deviation amount between the first tomographic image and the second tomographic image is measured by comparing the sector-shaped region of the first tomographic image and the sector-shaped region of the second tomographic image. X-ray CT apparatus as described in any one of these.   5. The X-ray CT apparatus according to claim 4, wherein the positional deviation amount is obtained from a similarity ratio between the fan-shaped area of the first tomographic image and the fan-shaped area of the second tomographic image.   5. The X-ray CT apparatus according to claim 4, wherein the positional deviation amount is obtained by correlation calculation between a sector area of the first tomogram and a sector area of the second tomogram.   The tomographic image position correction unit performs a weighted addition so that each sector area overlaps the sector area adjacent to each other and overlaps the sector areas, and the overlapping areas are continuously joined to each other. 6. The X-ray CT apparatus according to claim 4, wherein processing is performed.   The tomographic image position correction unit identifies pixels on a line segment extending radially from the center of gravity in each of the first tomographic image and the second tomographic image, and pixels on the line segment of the first tomographic image 4. The positional shift amount between the first tomographic image and the second tomographic image is obtained by comparing a value with a pixel value on the line segment of the second tomographic image. X-ray CT apparatus as described in any one of these.   9. The positional shift amount is obtained by calculating a correlation between a pixel value on the line segment of the first tomographic image and a pixel value on the line segment of the second tomographic image. X-ray CT system.   The X-ray CT apparatus according to claim 6, wherein the displacement amount is an enlargement / reduction ratio calculated using a half-value width of a profile obtained as a result of the correlation calculation.   11. The X-ray according to claim 1, wherein the tomographic image position correction unit displays the first tomographic image and the second tomographic image in a superimposed manner for comparison. CT device.   The tomographic image position correction unit performs coordinate conversion on at least one of the first tomographic image and the second tomographic image based on the amount of positional shift. X-ray CT apparatus according to item.   The X-ray CT apparatus according to claim 12, wherein the coordinate transformation is affine transformation or multi-dimensional coordinate transformation.