JP2012034972A - X-ray ct device, and signal processing program of the device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the image quality by detecting a motion of a photographing object and performing reconstruction according to the magnitude of the motion.SOLUTION: The X-ray CT device includes: an X-ray detector which is disposed to face the X-ray tube, the X-ray detector detecting the X-ray and outputs an X-ray detection signal; a reconstruction means which reconstructs a CT image based on the X-ray detection signal; and a weighting means which sets a weighting function associated with coordinate information of the CT image based on a difference between a first CT image reconstituted based on the X-ray detection signal detected in a first time area and a second CT image reconstituted based on the X-ray detection signal detected in a second time area. The reconstruction means reconstructs a third CT image based on the X-ray detection signal weighted according to the weighting function.

Description

  The present disclosure relates to an X-ray CT apparatus that captures a CT image of a subject, and a signal processing program for the X-ray CT apparatus.

  An X-ray CT apparatus irradiates a subject with X-rays from various angles, detects X-rays transmitted through the subject, and generates a CT image of the subject. Generation of tomographic images or three-dimensional volume data (hereinafter simply referred to as scanning) by this X-ray CT apparatus is performed while rotating an X-ray tube and an X-ray detector attached to the inside of an apparatus called a gantry. This is performed by irradiating the subject placed on the opening of the gantry with X-rays.

  In the CT image generated by such scanning, artifacts generated by the movement of the subject's body and internal organs are reflected. In order to reduce this artifact, the X-ray detection signal is weighted according to the collection time of the X-ray detection signal (X-ray projection data), and the CT image is reconstructed using the weighted X-ray detection signal Is disclosed.

JP 2003-199740 A

  In the conventional X-ray CT apparatus described above, the time resolution is improved by simply performing weighting according to the collection time regardless of whether the subject's body or internal movement has occurred. Although the effect of the movement of the subject can be reduced by improving the time resolution, the time resolution is in a trade-off relationship with the image quality, so the CT image is affected by noise as the time resolution is improved. Will get worse.

  By the way, when looking at the object being scanned, the magnitude of the movement and the position where the movement occurs vary from person to person, and changes variously depending on the imaging position of the CT image. For example, when imaging the abdomen of a subject, the movement of the spine is small compared to the intestinal tract that causes movement due to peristalsis. Artifacts due to movement are less likely to appear in this small area of movement. However, since the weighting in the conventional X-ray CT apparatus described above is also applied to such a small-motion region where there is little need for artifact reduction, there is a problem that the image quality of the CT image is unnecessarily deteriorated. there were.

  Therefore, in the present disclosure, the image quality is improved by detecting the movement of the photographing target and performing reconstruction according to the magnitude of the movement.

  In order to solve the above problems, an X-ray CT apparatus according to an embodiment includes an X-ray tube that irradiates X-rays and an X-ray that is disposed to face the X-ray tube, detects the X-rays, and outputs an X-ray detection signal. A detector, reconstruction means for reconstructing a CT image based on the X-ray detection signal, and a first CT image reconstructed based on the X-ray detection signal detected in the first time domain Weighting for setting a weighting function associated with coordinate information in the CT image based on a difference from the second CT image reconstructed based on the X-ray detection signal detected in the second time domain And the reconstruction means reconstructs a third CT image based on the X-ray detection signal weighted according to the weighting function.

  In order to solve the above-described problem, an X-ray CT apparatus according to an embodiment detects an X-ray and outputs an X-ray detection signal that is disposed opposite to the X-ray tube that irradiates X-rays. X-ray detector, reconstruction means for reconstructing a CT image based on the X-ray detection signal, first X-ray detection signal detected at a first time, and the first X-ray detection signal Weighting means for setting a weighting function associated with the coordinate information in the CT image based on the difference from the second X-ray detection signal detected at the second time that passes through the X-ray trajectory at And the reconstruction means reconstructs a third CT image based on the X-ray detection signal weighted according to the weighting function.

  In order to solve the above problems, an X-ray CT apparatus according to an embodiment detects an X-ray and outputs an X-ray detection signal that is disposed opposite to the X-ray tube that irradiates X-rays. In an X-ray CT apparatus comprising an X-ray detector and a reconstruction means for reconstructing a CT image based on the X-ray detection signal, reconstruction based on the X-ray detection signal detected in the first time domain The coordinate information in the CT image is associated with the first CT image based on the difference between the first CT image and the second CT image reconstructed based on the X-ray detection signal detected in the second time domain. A time resolution changing means for setting the set time resolution function, wherein the reconstruction means reconstructs a CT image based on the X-ray detection signal in which the time resolution function is set. .

  In order to solve the above problem, the signal processing program of the X-ray CT apparatus of the embodiment reconstructs the first CT image based on the X-ray detection signal detected in the first time domain; A step of reconstructing the second CT image based on the X-ray detection signal detected in the second time domain, and calculating a difference between the first CT image and the second CT image, Based on the step of setting a weighting function associated with the coordinate information in the CT image, the step of weighting the X-ray detection signal according to the set weighting function, and the weighted X-ray detection signal And reconstructing a third CT image.

The block diagram which shows the structure of the X-ray CT apparatus which concerns on an embodiment. Sectional drawing which shows the structure of the rotary body which concerns on an embodiment. The figure which shows the weighting function which concerns on an embodiment. FIG. 6 is another diagram showing a scanning process according to an embodiment. The figure which shows the state which completed the scan which concerns on embodiment. The figure which shows a mode that the reference image which concerns on an embodiment is reconfigure | reconstructed. The figure which shows a mode that another reference image which concerns on an embodiment is reconfigure | reconstructed. The figure which shows a mode that the motion amount map which concerns on an embodiment is produced | generated. The figure which shows a mode that the weighting function which concerns on an embodiment is produced | generated. The figure which shows a mode that another weighting function which concerns on an embodiment is produced | generated. The figure which shows a mode that another weighting function which concerns on an embodiment is produced | generated. The figure which shows another example of the weighting function which concerns on an embodiment. The flowchart which showed the production | generation process of CT image which concerns on embodiment. The figure which shows the process of the helical scan which concerns on an embodiment. The figure which shows another method of reconstructing the reference image which concerns on an embodiment.

(Configuration of X-ray CT apparatus 1)
Each embodiment will be described below with reference to the drawings. FIG. 1 is a block diagram showing an internal configuration of an X-ray CT apparatus 1 according to each embodiment.

  The control unit 100 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory). The control unit 100 includes a scan control unit 102, a reconstruction processing unit 103, an image processing unit 104, a display unit 106, a storage unit 107, and an input unit 108. The control unit 100 performs overall control of the X-ray CT apparatus 1 by processing signals supplied from the respective units and generating various control signals and supplying them to the respective units.

  The scan control unit 102 outputs an X-ray beam irradiation signal for irradiating the X-ray tube 301 with X-rays based on scan parameters input from the input unit 108 or the like when the X-ray CT apparatus 1 performs scanning. Output to the tube controller 201. Further, the scan control unit 102 rotates the rotating body 300 based on the scan parameters such as the slice thickness, the rotation speed, and the size of the scan area input from the input unit 108 and the like, and the first collimator 401 and the second collimator 401. A gantry drive signal for moving the collimator 430 is output to the gantry drive control unit 202. Further, the scan control unit 102 outputs a bed driving signal for moving the couchtop 500 to the bed driving control unit 203 based on scan parameters such as a scan position input from the input unit 108 and the like and a bed movement instruction.

  The reconstruction processing unit 103 performs back projection processing on the X-ray detection signal output when the X-ray detector 302 detects X-rays, and generates a CT image. When generating the CT image, the reconstruction processing unit 103 outputs the CT image to the image processing unit 104.

  The CT image generated by the reconstruction processing unit 103 is divided into a medical CT image used for diagnosis and a reference image used auxiliary for generating a medical CT image. In the following embodiments, a medical CT image that is a CT image and a reference image will be described separately. A medical CT image generation process, which will be described later, is generally performed as follows. First, the reconstruction processing unit 103 reconstructs reference images 810 and 811 based on the X-ray detection signal detected by the X-ray detector 302 at the time of scanning, and generates a motion map 820 by combining the reference images. . Then, the reconstruction processing unit 103 calculates a weighting function 830 based on the motion map 820, and reconstructs a medical CT image using the X-ray detection signal weighted along the weighting function 830. The reconstruction processing of the reference images 810 and 811, the generation processing of the motion map 820, the setting processing of the weighting function 830, and the reconstruction processing of the medical CT image using weighting will be described in detail later.

  The image processing unit 104 extracts a medical CT image having an arbitrary cross section from a plurality of medical CT images output from the reconstruction processing unit 103 based on an instruction input from the input unit 108, or a plurality of medical CT images. Are combined and converted into image data such as three-dimensional image data subjected to rendering processing. The image processing unit 104 outputs the extracted medical CT image and the converted three-dimensional image data to the display unit 106 or the storage unit 107.

  The display unit 106 is configured by a liquid crystal display, for example, and displays the image data output from the image processing unit 104. In addition, an operation screen for operating the X-ray CT apparatus 1 and a scan parameter input by the input unit 108 are displayed.

  The storage unit 107 is configured by combining a storage medium such as a ROM, a RAM, a flash memory that is an electrically rewritable and erasable nonvolatile memory, and an HDD (Hard Disc Drive). The storage unit 107 stores the image data output from the image processing unit 104, the weighting function 830 used for the reconstruction processing of the reconstruction processing unit 103, and stores various applications executed by the CPU of the control unit 100. .

  The input unit 108 includes, for example, a touch panel display, mechanical buttons, and the like, and receives an input made by the user of the X-ray CT apparatus 1 to the input unit 108. When the input unit 108 receives an input of a scan parameter, a scan start / stop instruction, a bed movement instruction, or the like performed by the user, the input unit 108 outputs an input signal according to the instruction performed by the user. The control unit 100 executes various processes according to the input signal output from the input unit 108.

  The X-ray tube control unit 201 receives the X-ray beam irradiation signal output from the scan control unit 102 and applies a high voltage for irradiating the X-ray tube 301 with X-rays. The application of the high voltage is performed along X-ray parameters specified by the X-ray beam irradiation signal, and the X-ray parameters specify parameters such as tube voltage, tube current, and X-ray pulse width.

  The X-ray tube 301 receives the high voltage applied from the X-ray tube control unit 201 and irradiates the X-ray detector 302 with X-rays. At this time, the X-ray tube 301 irradiates X-rays having a shape spreading in a fan shape in the column direction so that X-rays enter each column of the X-ray detector 302 configured to have a plurality of columns.

  The X-ray detector 302 detects the X-rays emitted from the X-ray tube 301 and outputs this to the reconstruction processing unit 103 as an X-ray detection signal. The X-ray detector 302 detects an incident X-ray and outputs an X-ray detection element that outputs an electric signal corresponding to the detection amount in the channel direction (the arrow direction in FIG. 2 or the rotation direction of the rotating body 300). A plurality of columns are arranged side by side. More specifically, the X-ray detector 302 performs X-ray along the channel direction of the subject P or along the direction (arrow direction in FIG. 2) perpendicular to the body axis direction (z direction in FIG. 1). A channel is configured by arranging detection elements, and a plurality of X-ray detection elements are arranged in a two-dimensional manner by arranging a plurality of channels of the X-ray detection elements along the column direction (z direction in FIG. 1). Is done. Hereinafter, the arrow direction in FIG. 2 is simply referred to as a channel direction, and the z direction in FIG. 1 is simply referred to as a column direction.

  The gantry drive control unit 202 receives the gantry drive signal output from the scan control unit 102, moves the first collimator 401 and the second collimator 430, and rotates the rotating body 300. These movements and rotations are controlled based on scan parameters such as a rotation speed and a scan range specified by the input unit 108.

  The first collimator 401 is a plate made of a substance that shields X-rays such as lead and tungsten. A plurality of first collimators 401 are provided so as to block the irradiation direction of the X-ray tube 301 and block a part of the X-rays irradiated from the X-ray tube 301. The X-rays emitted from the opening of the first collimator 401 spread in a fan shape toward the column direction and enter the X-ray detector 302 as indicated by the dotted line in FIG. A drive motor (not shown) is attached to the first collimator 401, and the first collimator 401 moves in the ± z direction in FIG. 1 to change the region that blocks X-rays. Similar to the first collimator 401, a second collimator 430 is provided so as to block the irradiation direction of the X-ray tube 301 in order to control the spread angle with respect to the channel direction. FIG. 2 is a cross-sectional view of the rotator 300 viewed from the xy plane. As shown by the dotted line in FIG. 2, the X-ray blocked by the first collimator 401 spreads in a fan shape with respect to the channel direction and enters the X-ray detector 302. A drive motor (not shown) is attached to the second collimator 430, and the second collimator 430 moves in the ± y direction in FIG. 2 to change the region blocking X-rays.

  The rotating body 300 is a cylindrical device built in the gantry 3. The rotating body 300 is attached with members such as an X-ray tube 301, a first collimator 401, a second collimator 430, and an X-ray detector 302. A rotary motor is attached to the rotator 300, and when the X-ray CT apparatus 1 performs scanning, the rotator 300 receives the gantry drive instruction signal output from the gantry drive control unit 202 and drives the rotator to A rotational motion about the body axis of the specimen P is performed. As the rotator 300 rotates, each member attached to the rotator 300 also performs a rotational motion around the body axis of the subject P.

  The top plate 500 is a plate-like member on which the subject P can be laid down. The top plate 500 is supported by the top plate drive unit 501, and the top plate drive unit 501 described later moves the top plate 500 along the longitudinal direction of the top plate 500 (the z-axis direction in FIG. 1).

  The top plate driving unit 501 is a device for moving the top plate 500, which is configured by a motor (not shown) or the like. The top plate driving unit 501 is configured by attaching a belt connected to a motor to the top plate 500, for example. The position of the top plate 500 moves in conjunction with the rotation of the motor of the top plate drive unit 501. The top plate drive unit 501 receives the top plate drive signal output from the scan control unit 102 and moves the top plate 500 along the longitudinal direction of the top plate 500 (z-axis direction in FIG. 1).

(Scan process flow)
When the X-ray CT apparatus 1 performs scanning, the X-ray tube 301 first irradiates X-rays from various angles toward the X-ray detector 302 while the rotator 300 rotates once. The X-ray detector 302 outputs an X-ray detection signal to the reconstruction processing unit 103 according to the irradiation angle, and the reconstruction processing unit 103 uses the medical CT image based on the X-ray detection signal output from the X-ray detector 302. Can be reconfigured.

  Here, when a single medical CT image is generated using the X-ray detection signals collected while the rotator 300 rotates once, it takes time for the rotator 300 to rotate once to collect data. It will be. Since the rotation of the rotating body 300 takes a certain time, for example, 0.5 seconds, a part of the tissue of the subject P may move until the rotating body completes one rotation ( Hereinafter, a region where tissue movement occurs is simply referred to as a movement region). The movement of the tissue of the subject P is caused by, for example, peristalsis of the internal organs, respiration of the subject P, or movement of muscles of the subject P.

  Here, consider the process in which the reconstruction processing unit 103 calculates the pixel value of the pixel corresponding to the moving region by the reconstruction process. As described above, during scanning, the X-ray tube 301 emits X-rays while changing the irradiation angle by the rotation of the rotating body 300. Then, the reconstruction processing unit 103 reconstructs the moving area using the X-ray detection signal transmitted through the moving area among the irradiated X-rays. At this time, the amount of X-ray absorption in the moving region changes momentarily as the tissue moves, so the value of the X-ray detection signal transmitted through the moving region also changes depending on the X-ray detection time. For this reason, when reconstruction is performed using the detection signal whose value has changed, the pixel value of the moving region of the generated medical CT image changes as compared to the case where there is no movement. By changing the value of the pixel, it appears that something that does not originally exist in the subject P is reflected. An object that does not exist in the subject P and appears in the medical CT image is called an artifact, and the artifact hinders diagnosis using the medical CT image.

  Artifacts caused by the movement of the tissue of the subject P can be suppressed by increasing the time resolution when generating a medical CT image. In the X-ray CT apparatus 1 of the present disclosure, the time resolution at the time of generating a medical CT image is changed by weighting the X-ray detection signal used for the reconstruction of the medical CT image.

  FIG. 3 shows a weighting function 830 that represents a weighting profile for the X-ray detection signal. The horizontal axis of the weighting function 830 plots the imaging time and the view corresponding to the position of the X-ray tube 301 (the position of the X-ray tube 301), while the vertical axis plots the weight value. As an example, the weighting function 830 shown in FIG. 3 outputs an X-ray detection signal for 900 views when the X-ray tube 301 outputs an X-ray detection signal for 900 views while rotating around the subject P once. Is set as a weight. FIG. 3A shows an example of the weighting function 830 when weighting is performed, and FIG. 3B shows an example of the weighting function 830 when weighting is not performed.

  The medical CT image is backed up using a set of X-ray detection signals (hereinafter simply referred to as raw data 800) collected by the reconstruction processing unit 103 while rotating the X-ray irradiation position around the subject P. It can be obtained by performing reconstruction processing based on the projection method. In the medical CT apparatus 1 of the present disclosure, when the reconstruction processing unit 103 performs the reconstruction process, the time resolution is changed by applying the weight set by the weighting function 830 to the raw data 800.

  For example, X-ray projection data (hereinafter simply referred to as a CT value) for a certain pixel in a medical CT image is obtained by a value obtained by performing line integration of 900 views of X-ray detection signals passing through a corresponding path. Based on the weighting function 830, the reconstruction processing unit 103 weights the X-ray detection signals for 900 views. When the weighting function 830 shown in FIG. 3A is used, the weight value for the X-ray detection signal around the 449 view is increased. On the other hand, the weight for the X-ray detection signal around the 0 view or 900 view corresponding to the view having an increased weight value and passing through the same path as the X-ray path in the view having the increased weight value is 449 views. The weight value is decreased to compensate for the increased weight value in the vicinity.

  Accordingly, the X-ray detection signal around the 449 view with the increased weight value is dominant in calculating the CT value. On the other hand, the influence of the X-ray detection signals around the 0 view or 900 view on the CT value is relatively small. Since a certain time is required for the rotator 300 to make one rotation, the time when the X-ray detection signal in the 0 view is acquired and the time when the X-ray detection signal in the 900 view are acquired are due to the rotation time of the rotator 300. Time interval to be freed. For this reason, when a series of X-ray detection signals are collected in a state in which body motion occurs in the subject P, the X-ray detection signals around the 0 view and the 900 view that have a wide time interval pass through the same path, for example. Even if it is an X-ray detection signal, the value will be detected differently. By reducing the weight values of the X-ray detection signals around the 0 view and the 900 view, it is possible to reduce the influence of fluctuations in the X-ray detection signals due to body movements on the CT value. By weighting the X-ray detection signal, the CT value can be calculated mainly using the X-ray detection signal collected in a short time region. That is, the weighting for the X-ray detection signal can be regarded as substantially the same as improving the time resolution at the time of generating the medical CT image.

  By the way, when the reconstruction is performed by weighting as shown in FIG. 3A and when the X-ray detection signal is not weighted, that is, uniform weighting as shown in FIG. Compare with the case of reconstruction. When the weighting of FIG. 3A is performed, since the weight value in some X-ray detection signals is large, the influence of random noise mixed in the X-ray detection signals on the CT value becomes large. On the other hand, when the weighting of FIG. 3B is performed, since the weight value in the X-ray detection signal is uniform, random noise mixed in each X-ray detection signal is added to the X-ray detection signal. Therefore, the influence of noise on the CT value is reduced. In other words, the improvement of the time resolution by weighting has a greater effect on the quality of the medical CT image due to noise. Therefore, in order to obtain a medical CT image useful for diagnosis, the temporal resolution is improved to reduce the artifacts for the region with a large amount of movement, while the temporal resolution is decreased for the region with a small amount of movement. Therefore, it is necessary to improve the image quality.

  Therefore, in this embodiment, the distribution of the moving area on the subject P is detected, and reconstruction is performed by changing the time resolution for each area according to the moving amount. Specifically, a set of X-ray detection signals collected by the reconstruction processing unit 103 by scanning, that is, reference images 810 and 811 having different imaging times are reconstructed from the raw data 800, and the difference between the reference images 810 and 811 is reconstructed. As a result, a weight map 820 is generated in which the moving region indicates the distribution and the moving amount in any region in the subject P. Further, a weighting function 830 is set according to the amount of movement in the moving area read from the weight map 820. The weighting function 830 is a function for associating the amount of motion with temporal resolution, and weighting that improves the temporal resolution for regions with a large amount of motion and decreases the temporal resolution for regions with a small amount of motion. A function 830 is assigned to each pixel of the medical CT image. This assignment of the weighting function for each pixel may be rephrased as a process of associating the weighting function with the coordinate information of the medical CT image. When the weighting function 830 is set, the reconstruction processing unit 103 reconstructs a medical CT image according to the weighting function 830 set for each region. As a result, it is possible to reduce artifacts by increasing the temporal resolution for regions with a large amount of motion and performing reconstruction, while reducing the temporal resolution for regions with a small amount of motion and little artifacts. It is possible to generate a medical CT image with improved image quality.

  More specifically, the X-ray CT apparatus 1 described in this embodiment generates a medical CT image in which artifacts are suppressed by performing processing substantially in the following order.

(1) Collection of raw data 800 (2) Reconstruction of reference 810 and 811 images (3) Generation of weight map 820 (4) Setting of weighting function 830 (5) Reconstruction of medical CT image A process flow of the X-ray CT apparatus 1 when generating a medical CT image will be described.

(1) Collection of Raw Data 800 When the X-ray CT apparatus 1 starts scanning, the X-ray tube 301 first irradiates the subject P with X-rays while the rotating body 300 is rotated, and the X-ray detector 302 collects raw data 800. Here, the raw data 800 refers to a set of X-ray detection signals collected by the X-ray detector 302 for each irradiation angle (hereinafter, the irradiation angle of the X-ray tube 301 is simply referred to as a view). One raw data 800 corresponds to a set of X-ray detection signals collected by one column of X-ray detection elements during one rotation of the X-ray tube 301, and the raw data 800 is defined by two arguments, a channel and a view described later. It is stored in the storage unit 107 as defined two-dimensional array data. In the following embodiments, for the sake of simplicity, description will be made with attention paid to a single row of X-ray detection elements in the X-ray detector 302.

  4 and 5 are diagrams showing raw data 800 collected by the X-ray detector 302 as the rotating body 300 rotates. 4 and 5, raw data 800 that is two-dimensional array data is shown as a two-dimensional image having different brightness according to the intensity of the X-ray detection signal. The horizontal axis of the image of the raw data 800 corresponds to the channel, and the vertical axis corresponds to the view. Further, in this embodiment, as an example, each column of the X-ray detector 302 is described as being configured by arranging 512 X-ray detection elements (corresponding to the description of 0ch-511ch in FIG. 4) (hereinafter referred to as “the description”). The order in which the X-ray detection elements are arranged in the channel direction is simply referred to as a channel). The first collimator 401 and the second collimator 430 do not control the X-ray irradiation range, and the X-rays irradiated from the X-ray tube 301 are incident on all X-ray detection elements. Naturally, the number of X-ray detection elements in each column may be adjusted, and the X-ray irradiation range may be appropriately narrowed by controlling the first collimator 401 and the second collimator 430.

  Hereinafter, collection of the raw data 800 by the X-ray detector 302 will be described. First, when the X-ray tube 301 emits X-rays, the X-rays pass through the subject P and enter the X-ray detector 302. The intensity of the X-rays when entering the X-ray detector 302 changes by passing through the subject P, and X-rays having different intensities are incident on the arranged X-ray detection elements for each channel. When attention is paid to a specific column of X-ray detection elements constituting the X-ray detector 302, 512 X-ray detection signals are output from this column for each detection. The X-ray tube 301 and the X-ray detector 302 continue X-ray irradiation and detection while changing the X-ray irradiation angle by the rotation of the rotating body 300. If the X-ray detector 302 has detected 125 views by the time it reaches the state of FIG. 4, the raw data 800 in a certain column at this time has 512 X-ray detection signals for 125 views per view. It will be. Further, the rotation of the rotating body 300 and the detection of X-rays by the X-ray detector 302 are continued, and when the rotating body 300 rotates once as shown in FIG. 5, the X-ray CT apparatus 1 collects raw data 800 for one scan. Exit. Assuming that the X-ray detector 302 detects 900 views before the rotation of the rotating body 300, the raw data 800 in a certain column at this time has 512 X-ray detection signals for 900 views per view. It will be. The raw data 800 collected by the above operation is output to the reconstruction processing unit 103, and the reconstruction processing unit 103 generates reference images 810 and 811 described later using the raw data 800.

(2) Reconstruction of reference images 810 and 811 The reconstruction processing unit 103 reconstructs the reference images 810 and 811 based on the raw data 800 described above. The reference images 810 and 811 are CT images reconstructed from a part of the data area of the raw data 800, or a set of reconstructed data. The reference images 810 and 811 are obtained by generating a tomographic image based on the CT value obtained by back projecting the raw data 800. For example, two reference images 810 and 811 are generated from the raw data 800 obtained from one scan. The reconstruction process is performed so that the data area of the raw data 800 to be reconstructed of the reference image 810 and the data area of the raw data 800 to be reconstructed of the reference image 811 include data areas having different acquisition times of X-ray detection signals The unit 103 sets a data area to be reconfigured.

  FIG. 6 shows how the reference image 810 is reconstructed. The reconfiguration process will be specifically described below. First, the reconstruction processing unit 103 performs a filtering process on the raw data 800 for enhancement of the reconstructed CT image as preprocessing for reconstruction. This filtering process is performed, for example, by filtering a high-frequency component included in the raw data 800. Although the description is omitted here for simplicity, various correction processes such as X-ray intensity correction and beam hardening correction may be performed on the raw data 800 in order to improve image quality. When the reconstruction processing unit 103 finishes applying the filter processing, the reconstruction processing unit 103 extracts a set of X-ray detection signals included in a data area in which the raw data 800 exists. FIG. 6 shows, as an example, a case where X-ray detection signals included in the data area from 1 view to 573 view, which is the first half of the data area in the raw data 800, are extracted. The reconstruction processing unit 103 performs reconstruction processing based on the back projection method on the extracted X-ray detection signal, and generates a reference image 810 in which CT values are mapped in a two-dimensional space.

  FIG. 7 shows how the reference image 811 is reconstructed. The content of the reconstructing process of the reference image 811 is the same as the reconstructing process of the reference image 810 described above. The reference image 811 is reconstructed using, for example, an X-ray detection signal included in a data area from 326 views to 900 views, which is the latter half area of the raw data 800.

  Here, if the reference images 810 and 811 are reconstructed based on the data areas as shown in FIGS. 6 and 7, there will be a shift in the collection time of the X-ray detection signal that is the basis of each image. Specifically, for example, assuming that the rotator 300 makes one rotation over 0.50 seconds, the X-ray detection signals from the 1st view to the 573th view constituting the reference image 810 are 0. It is collected in the time domain until 32 seconds elapse. On the other hand, the X-ray detection signals from 326 views to 900 views constituting the reference image 811 are collected in the time domain from the time when 0.18 seconds have elapsed from the start of scanning until the time when 0.50 seconds have elapsed. It is a thing.

  As described above, when there is a moving region in which the movement of the tissue occurs in the subject P, the amount of X-ray absorption in the moving region changes every moment due to the movement of the tissue. Since the reference image 810 and the reference image 811 have different X-ray detection signal acquisition times, the CT values in the moving region appear as different values on the reference images. Therefore, it is possible to detect in which part the moving region on the subject P is distributed by detecting the region where the CT value changes on each reference image. The reconstruction processing unit 103 detects a moving area on the subject P by generating a weight map 820 using the reconstructed reference images 810 and 811.

  In the present embodiment, the case where two reference images are reconstructed has been described, but a process of generating three or more reference images may be performed. In addition, although it has been described that the two reference images are reconstructed using the first half area and the second half area in the data area of the raw data 800, reconstruction may be performed using different data areas. For example, the reference image 810 is generated based on the data region from 100 views to 673 views in the data region of the raw data 800, and the reference image 811 is generated based on the data region from 200 views to 773 views. Of these, reconstruction may be performed using two sets of data in the intermediate area. Further, two reference images may be reconstructed from two raw data 800 collected at different times. In addition, a reference image may be generated from a data area that spans two raw data 800 collected continuously in time. For example, one reference image is generated based on data from 700 views to 900 views of a certain raw data 800 and data from 1 view to 473 views of the raw data 800 collected immediately after the raw data 800. It doesn't matter.

(3) Generation of Weight Map 820 When the reconstruction processing unit 103 reconstructs the reference images 810 and 811, it detects a difference between the two reference images 810 and 811 and generates a weight map 820. FIG. 8 shows how the weight map 820 is generated. Specifically, the difference is detected by converting the difference between the CT values mapped to the respective pixels of the two reference images 810 and 811 to an absolute value, and using the absolute value as a pixel value, the same x-axis as in the reference images 810 and 811 This is done by mapping to a two-dimensional space consisting of two axes of the y axis.

  As described above, the difference in CT value between the reference image 810 and the reference image 811 is caused by a change in the amount of X-ray absorption in the moving region. The difference amount of the CT value becomes a larger pixel value and appears on the weight map 820 as the change in the X-ray absorption amount in the movement region is larger, that is, as the movement amount in the movement region is larger. Therefore, the distribution of the moving area in the subject P and the amount of movement are substantially mapped on the weight map 820.

  Since noise components are mixed in the reference images 810 and 811, the weight map 820 includes pixels that do not reflect the movement of the moving area. In order to remove such pixels, the reconstruction processing unit 103 performs a smoothing process on the weight map 820. The smoothing process is a process of correcting a pixel value mapped to a pixel at a certain coordinate and changing the pixel value with a peripheral pixel so as to be smooth. The smoothing process is performed by applying various image filters such as a low-pass filter, a moving average filter, or a median filter to the weight map 820. Noise can be removed from the weight map 820 by performing the smoothing process.

  In addition, the reconstruction processing unit 103 performs normalization processing on each pixel of the weight map 820 that has been subjected to smoothing processing. The normalization process is a process of converting each pixel value in the weight map 820 into a value in a range suitable for calculation such as setting of a weighting function 830 described later. The normalization process is performed, for example, by extracting a pixel value having a maximum value in the weight map 820 and dividing the pixel value of each pixel by this maximum value. By this calculation, each pixel value in the weight map 820 is normalized to a value in the range of 0 to 1.

  The reconstruction processing unit 103 generates a weight map 820 that has been subjected to smoothing processing and normalization processing by the above processing. The reconstruction processing unit 103 sets a weighting function 830 based on each pixel value of the weight map 820.

(4) Setting of Weighting Function 830 When the reconstruction processing unit 103 generates the weight map 820, the weighting function 830 is set for each pixel based on the motion amount value mapped to each pixel of the weight map 820. . In a medical CT image reconstruction process to be described later, the reconstruction processing unit 103 performs a reconstruction process using all X-ray detection signals from one view to 900 views. This reconstruction processing is performed by adding a weight based on the weighting function 830 to the X-ray detection signal and then reconstructing a medical CT image by the back projection method. Here, the weighting function 830 is a function for weighting 900 X-ray detection signals extracted from the raw data 800. The reconstruction processing unit 103 calculates a CT value by weighted addition using a weighting function 830 described later, and performs a medical CT image reconstruction process.

  The reconstruction processing unit 103 sets a weighting function 830 having a different shape for each pixel based on the amount of motion mapped to each pixel in the weight map 820. There are at least three types of weighting functions 830, and weighting functions 830 having different shapes are stored in the storage unit 107 in association with the amount of motion. In this case, the reconstruction processing unit 103 sets the weighting function 830 by reading out the weighting function 830 associated with the motion amount from the storage unit 107. Although the shape of the weighting function 830 affects the time resolution described later, by preparing a plurality of weighting functions 830 having different shapes, it is possible to suppress the occurrence of unnatural pixels with greatly different time resolution. Instead of storing the weighting function 830 itself in the storage unit 107, a coefficient for setting the shape of the weighting function 830 may be stored in the storage unit 107 in association with the amount of motion. In this case, the reconstruction processing unit 103 reads the coefficient associated with the motion amount from the storage unit 107, and sets the weighting function 830 by calculating the weighting function 830 based on the coefficient.

  FIG. 9 shows an example of the weighting function 830 used when reconstructing a pixel at a certain coordinate [x1, y1] in the medical CT image. The weighting function 830 has a multi-stage value ranging from 0 to 2, is set so that the weight value is switched continuously, and is set to have a weight of three or more levels of 0, 1, and 2. The In addition, the weighting function 830 increases the weight value as it approaches the 449th view that is the center of the data area of the raw data 800, and approaches the maximum value in the 449th view. On the other hand, the weight value decreases as it approaches the first or 900th view that is the end of the data area, approaches 0, and is set to take the minimum value in the 1st and 900th views. For example, when the weighting function 830 as shown in FIG. 9 is set, the X-ray detection signal belonging to the 449th view is treated as having a detection amount nearly doubled at the time of weighted addition, whereas the first and 900th views The X-ray detection signals belonging to the views are weighted so as to have a detection amount close to 0 at the time of weighted addition. That is, the shape of the weighting function 830 is, for example, a shape in which the weight value for the view at the center of the data region is increased while the weight value for the view at the end of the data region is decreased to compensate for this increase. More specifically, the sum of the weight value of a view and the weight value of the view passing through the same path as this view is 1, and the weighting function 830 compensates so that the weight value for each view is compensated. Is set.

  When the reconstruction processing unit 103 performs such weighting on the X-ray detection signal and performs the reconstruction process, the CT value of the pixel at a certain coordinate [x1, y1] is the central time among the collection times of the raw data 800. The X-ray detection signal close to (corresponding to the collection time of the 449th view) is dominantly determined, while the time away from the central time (corresponding to the collection time of the view closest to the first or 900th view). The degree of contribution of the X-ray detection signal close to the calculation of the CT value is small. This weighting process can be substantially regarded as a process for improving the time resolution.

  In this embodiment, the time resolution is adjusted by changing the shape of the weighting function 830 in accordance with the amount of motion detected from the weight map 820. Specifically, for a pixel with a large amount of motion on the weight map 820, the weight value at the time away from the central time is greatly decreased to increase the weight value around the central time in order to improve the time resolution. An increased weighting function 830 is set. On the other hand, in order to reduce the temporal resolution for pixels with a small amount of motion on the weight map 820, the amount of decrease in the value of the weight of the X-ray detection signal collected at a time away from the central time is reduced. A weighting function 830 in which the weight value of the peripheral X-ray detection signal is reduced is set.

  Hereinafter, a state in which the weighting function 830 is set based on the motion amount mapped to the weight map 820 is shown. The setting of the weighting function 830 performed by the reconstruction processing unit 103 is, for example, a weight value in proportion to the amount of motion detected by the weight map 820 (corresponding to “strength” in FIGS. 9, 10, and 11). This is performed by reading the weighting function 830 in which the width of the region for increasing the width and the width of the region for decreasing is changed from the storage unit 107.

  FIG. 10 shows a weighting function 830 when reconstructing a pixel at a certain coordinate [x2, y2]. Since the motion amount of the coordinates [x2, y2] is larger than the motion amount of the coordinates [x1, y1], the weighting function 830 is set to a shape that improves the time resolution compared to the coordinates [x1, y1]. Specifically, the weight of the region where the weight increases from 1 to 2 around the central time is wider than the weighting function 830 of the coordinates [x1, y1], and the weight of the region further away from the central time is increased. The weighting function 830 in which the width of the region where the value decreases from 1 to 0 is wider than the weighting function 830 of the coordinates [x1, y1] is read from the storage unit 107. As the weight value of the time away from the central time is decreased, the time resolution is substantially improved.

  FIG. 11 shows a weighting function 830 for reconstructing a pixel at a certain coordinate [x3, y3]. Since the amount of movement of the coordinates [x3, y3] is smaller than the amount of movement of the coordinates [x1, y1], the weighting function 830 has a shape that lowers the time resolution compared to the coordinates [x1, y1]. It is read from 107 and set. Specifically, in a region further away from the central time so that the width of the region where the weight value increases from 1 to 2 around the central time becomes narrower than the weighting function 830 of the coordinates [x1, y1]. A weighting function 830 is set in which the width of the area where the weight value decreases from 1 to 0 is narrower than the weighting function 830 of the coordinates [x1, y1]. As the amount of decrease in the weight value at the time away from the center time decreases, the time resolution substantially decreases.

  In the present embodiment, the weighting function 830 is stored in the storage unit 107 in advance, and the operation in which the reconstruction processing unit 103 reads and sets this from the storage unit 107 has been described. However, the weighting function 830 may be calculated by the reconstruction processing unit 103 in accordance with the amount of motion instead of being stored in advance. Also, in the present embodiment, as an example, the weighting function 830 is described in which the weight value for the center view of the data area increases while the weight value for the edge of the data area decreases. However, the area where the weight value of the weighting function 830 increases is not limited to the center of the data area. For example, the weight value for the view at the center of the corresponding data area is increased by increasing the weight value for the view at the end of the data area. A weighting function 830 with a reduced value may be used. Further, the area where the weight value is increased or decreased is not limited to the center or end of the data area, but a weighting function 830 in which an arbitrary area such as an intermediate area between the center and the end is increased or decreased may be used. Absent.

  In this embodiment, as an example of the weighting function 830, an example in which the weight value transitions from 0 to 1 and from 1 to 2 has been described. However, the weighting function 830 of the present embodiment is not limited to this, and as shown in FIG. 12A, the weight value is changed from 0.5 to 1 and from 1 to 1.5. The range in which the value changes may be changed as appropriate. Alternatively, a weighting function 830 having a different width in which the weight value changes according to the amount of motion detected from the weight map 820 may be used. In the weighting function 830 having the shape shown in FIG. 12A, when the amount of motion is large, the time resolution is substantially improved by using the weighting function 830 in which the weight value changes in the range of 0 to 2, for example. On the other hand, when the amount of motion is small, the time resolution can be lowered by using the weighting function 830 whose weight value changes in the range of 0.5 to 1.5, for example.

  Also, as shown in FIG. 12B, a weighting function 830 may be set for the X-ray detection signals collected over one rotation. FIG. 12B shows an example in which weighting is performed on X-ray detection signals for 1100 views as an example. In FIG. 12B, the weight value in 0 to 200 views is set to increase gently from 0 to 1. On the other hand, the weight value in 900 to 1100 views passing through the same path as 0 to 200 views gradually decreases from 1 to 0 so that the weight value of 0 to 200 views is compensated and the total value becomes 1. It is set as follows. When the amount of motion is small, weighting is performed on the X-ray detection signals from 0 view to 1300 views using the weighting function 830 in which the number of views that have exceeded is increased. A weighting function 830 is used in which the weight value increases in the range of 0 to 400 views, while the weight value decreases in the range of 900 to 1300 views. On the other hand, when the amount of motion is large, weighting is performed on the X-ray detection signals from 0 view to 1000 views using the weighting function 830 in which the number of excess views is reduced. A weighting function 830 is used in which the weight value increases in the range of 0 to 100 views, while the weight value decreases in the range of 900 to 1000 views. As described above, the time resolution can be changed by using the weighting function 830 in which the number of views exceeding the amount of motion and the range of the view whose value changes are different.

(5) Reconstruction of Medical CT Image When the reconstruction processing unit 103 sets the weighting function 830, it reconstructs a medical CT image according to the set weighting function 830. The medical CT image is reconstructed using X-ray detection signals included in all data regions in the raw data 800.

  The medical CT image reconstruction process will be described more specifically below. The reconstruction processing unit 103 extracts a set of X-ray detection signals included in the raw data 800 before starting reconstruction.

Next, the reconstruction processing unit 103 associates each pixel on the medical CT image with a cross section through which X-rays on the specimen P are transmitted. The reconstruction processing unit 103 calculates to which X-ray detection signal on the raw data 800 the X-ray detection signal of the X-ray that has transmitted the coordinates [x, y] on the medical CT image corresponds. An X-ray detection signal is extracted. In the reconstruction of the medical CT image, 900 X-ray detection signals included in the raw data 800 are extracted. Next, the reconstruction processing unit 103 reads a motion amount corresponding to a certain pixel [x, y] from the weight map 820 and determines a weighting function 830 corresponding to the motion amount. The reconstruction processing unit 103 weights a set of X-ray detection signals corresponding to a certain pixel [x, y] based on the weighting function 830 to change the value of the detection amount. The reconstruction processing unit 103 weights and adds the detection amount values of the 900 X-ray detection signals whose values have changed, and maps the obtained value to the pixel [x, y] in the medical CT image as a CT value. To do. Similar processing is performed for all pixels, and the reconstruction processing unit 103 reconstructs a medical CT image.

  As described in the section “(4) Setting the weighting function 830”, the weighting function 830 is set so that the time resolution is high in the region where the amount of motion is large on the subject P. By improving the time resolution, it is possible to reduce the influence due to the movement of the tissue and suppress artifacts appearing on the medical CT image. On the other hand, the weighting function 830 is set so that the time resolution is low in the region where the movement amount is small on the subject P. By reducing the time resolution, the CT value is substantially determined based on a larger number of X-ray detection signals, so that the influence of noise can be reduced and the image quality can be improved.

  As described above, the reconstruction of the medical CT image in this embodiment is performed by detecting the movement of the tissue for each pixel and adjusting the time resolution in accordance with the amount of movement. By performing reconstruction by improving the temporal resolution only for an area where the movement of the tissue is large and the artifact is large, it is possible to improve the image quality of the area other than the moving area while reducing artifacts generated in the moving area.

(Flow of medical CT image reconstruction process)
FIG. 13 is a flowchart showing the flow of medical CT image reconstruction processing in the present embodiment. Hereinafter, the flow of the reconstruction process will be described with reference to FIG. Note that when performing the reconstruction process of FIG. 13, it is assumed that the raw data 800 is collected and the weight map 820 is already generated by the X-ray detector 302.

  First, when the reconstruction processing unit 103 starts the reconstruction process (step 1000), the reconstruction processing unit 103 extracts a pixel [X, Y] having a medical CT image from the raw data 800 collected by the X-ray detector 302. Which channel on the raw data 800 the transmitted X-ray X-ray detection signal belongs to is calculated, and the corresponding view and channel X-ray detection signals are extracted (step 1001). Note that the function ch = CalcCH (X, Y, VIEW) in FIG. 13 is the raw data corresponding to the coordinates [X, Y] of the pixel on the medical CT image and the view number [VIEW]. This is a function for calculating a channel number [ch] on 800.

  Next, the reconstruction processing unit 103 reads the motion amount of the pixel at the coordinates [X, Y] (“strength” in FIG. 13) based on the weight map 820 (step 1002). A function of strength = StrengthMAP [X, Y] in FIG. 13 is a function for reading a value mapped to the coordinates [X, Y] on the weight map 820 as a motion amount [strength].

  Next, the reconstruction processing unit 103 sets a weighting function 830 based on the value of the read motion amount [strength], and based on the set weighting function 830, the weight of a certain view (“weight” in FIG. 13). A value is calculated (step 1003). Note that the function weight = CalcWeight (strength, VIEW) in FIG. 13 is obtained from the weighting function 830 when the motion amount [strength] and the view number [VIEW] read in step 1002 are given. ] Is a function for calculating

  Next, the reconstruction processing unit 103 weights and adds the weighted X-ray detection signal value to the CT value of the pixel at the coordinates [X, Y] of the medical CT image (Pixel [X, Y] in FIG. 13). (Step 1004). Note that the function Pixel [X, Y] = Ray [ch, VIEW] * weight + Pixel [X, Y] in FIG. 13 belongs to the channel number [ch] calculated in step 1001 and the given view number [VIEW]. The value of the X-ray detection signal (“Ray [ch, VIEW]” in FIG. 13) is read from the raw data 800, and the read X-ray detection signal value is multiplied by the weight value calculated in Step 1003. The function is a function for adding the value obtained by multiplication to the value Pixel [X, Y] mapped to the coordinates [X, Y] of the medical CT image.

  When the weighting addition of CT values is completed, the reconstruction processing unit 103 determines whether or not the processing according to Step 1001 to Step 1004 has been performed for all views (Step 1005). Note that “Nview” in FIG. 13 indicates the number of views obtained by one scan. When the reconstruction processing unit 103 determines that there is an unprocessed view (Yes in step 1005), the reconstruction processing unit 103 updates the view number [VIEW] (step 1006), and again in step 1001. Return to processing.

  When the reconstruction processing unit 103 determines that processing has been performed for all views (No in step 1005), the reconstruction processing unit 103 determines whether or not the processing according to steps 1001 to 1005 has been performed for all x coordinates. Determination is made (step 1007). Note that “MatrixX” in FIG. 13 indicates the maximum x-coordinate value in the medical CT image. When the reconstruction processing unit 103 determines that there is an x-coordinate that has not been processed (Yes in step 1007), the reconstruction processing unit 103 updates the view number [VIEW] and the x-coordinate value [X]. (Step 1008), the process returns to Step 1001 again.

  When the reconstruction processing unit 103 determines that processing has been performed for all x coordinates (No in Step 1007), the reconstruction processing unit 103 determines whether the processing according to Step 1001 to Step 1007 has been performed for all y coordinates. Is determined (step 1009). Note that “MatrixY” in FIG. 13 indicates the maximum y-coordinate value in the medical CT image. If the reconstruction processing unit 103 determines that there is an unprocessed y coordinate (Yes in step 1009), the reconstruction processing unit 103 determines that the view number [VIEW], the x coordinate value [X], and the y coordinate [Y] is updated (step 1010), and the process returns to step 1001 again. When the reconstruction processing unit 103 determines that processing has been performed for all y coordinates (No in step 1009), the reconstruction processing unit 103 ends the reconstruction of the medical CT image (step 1011).

  Through the above processing, the reconstruction processing unit 103 reconstructs a medical CT image. Note that the reconstruction process described in the present embodiment is not limited to this. For example, the order of the processes described in Step 1001 to Step 1004 may be appropriately changed, and Step 1005, Step 1007, and Step 1009 may be replaced. The order of the judgments described may be appropriately changed.

  In the above processing, attention is paid to a specific row of the X-ray detector 302, and a case where a medical CT image for one cross section is generated from a row of X-ray detection elements has been described. The same applies to the processing when the X-ray detection elements are arranged. By performing the above-described reconstruction process on the raw data 800 output from each column, a medical CT image having a plurality of cross sections can be generated.

(Generation of weight map 820 in helical scan)
The medical CT image reconstruction processing described in the previous embodiment has been described focusing on so-called conventional scanning in which scanning is performed with the top 500 fixed at a fixed position. However, the medical CT image reconstruction process described in the present embodiment is not limited to the conventional scan, and the top plate 500 is moved with respect to the gantry 3 along the longitudinal direction (z-axis direction in FIG. 14). It can also be applied to a helical scan that scans and collects data in a spiral.

  FIG. 14 shows a scan state in the helical scan. In the actual helical scan, the top 500 moves relative to the gantry 3, but FIG. 14 shows the gantry 3 relatively moved in FIG.

  When the X-ray CT apparatus 1 starts the helical scan, the X-ray tube 301 irradiates each row of the X-ray detector 302 so that the X-rays are incident on the X-ray detector 302 in a state where the rotating body 300 is rotated. The instrument 302 detects X-rays emitted from various angles. During the helical scan, the top plate 500 moves at a constant speed in the longitudinal direction, so that the z coordinate of the X-ray detection element that collects the X-ray detection signal changes with the imaging time. When the X-ray detector 302 moves the top plate 500 by a predetermined distance and finishes the helical scan, the X-ray detector 302 outputs the raw data 800 collected from each column to the reconstruction processing unit 103.

  When the reconstruction processing unit 103 reconstructs a medical CT image of a cross section, the reconstruction processing unit 103 outputs an X-ray detection signal transmitted through an area corresponding to the coordinates [x, y] of the medical CT image. Extract from each raw data 800. Then, a CT value is calculated by weighted addition of the extracted X-ray detection signals, and this CT value is mapped to coordinates [x, y] on the medical CT image. By performing this mapping for each coordinate on the medical CT image, the reconstruction processing unit 103 performs reconstruction.

  Here, in the example of FIG. 14, when attention is paid to the region corresponding to the coordinates [x, y] of the medical CT image, the X-ray detection signal transmitted through the region corresponding to the coordinates [x, y] is transmitted through the X-ray 910. Then, there will be one rotation until the X-ray 911 is transmitted. The same can be said for a medical CT image, and an X-ray detection signal for reconstructing a medical CT image of a certain cross section exists for one rotation. The X-ray detection signal for one rotation may include an X-ray detection signal interpolated by helical interpolation. For example, an X-ray detection signal at an intermediate position between two positions obtained by interpolating X-ray detection signals at two adjacent positions may be included.

  Therefore, by generating the reference images 810 and 811 and the weight map 820 shown in FIG. 6 and the like, it is possible to perform reconstruction processing in consideration of the moving region even in the helical scan. Specifically, when reconstructing a medical CT image of a certain section, the reconstruction processing unit 103 extracts an X-ray detection signal corresponding to the section to be reconstructed from each raw data 800. Further, the extracted raw data 800 is arranged in order of imaging time, and is classified into an X-ray detection signal belonging to the first half area of the imaging time and an X-ray detection signal belonging to the second half area. Note that this classification may be performed so that some X-ray detection signals overlap as shown in FIGS. 6 and 7.

  The reconstruction processing unit 103 reconstructs the reference images 810 and 811 from the X-ray detection signal belonging to the first half area and the X-ray detection signal belonging to the second half area, and calculates the weight by taking the absolute value of the difference between the CT values of the reference images. A map 820 is generated. When the reconstruction processing unit 103 generates the weight map 820, it reconstructs a medical CT image based on the processing described above with reference to FIG. With the above processing, it is possible to perform reconstruction in consideration of the moving region even in helical scanning in which scanning is performed while the top plate 500 is moved.

(Generation of weight map 820 using opposing X-ray detection signals)
In the above-described embodiment, as described with reference to FIGS. 6, 7, and 8, the reference images 810 and 811 are once reconstructed using X-ray detection signals belonging to a certain data area of the raw data 800. The weight map 820 is generated by taking the difference between the reconstructed reference images 810 and 811. However, the generation of the weight map 820 is not limited to this method.

  FIG. 14 shows another example of generating the weight map 820. The X-ray tube 301 and the X-ray detector 302 rotate while collecting X-rays and collecting X-ray detection signals, and when the rotation is completed, the scanning is finished. For this reason, for each X-ray detection signal, a pair of X-ray detection signals (hereinafter simply referred to as facing data) that enter the X-ray detector by following a trajectory opposite to the trajectory transmitted through the subject P. ) Will exist. FIG. 14 shows an example of a pair of X-ray detection signals. For example, an X-ray 900 incident on the center of the X-ray detector in the 573th view is an X-ray 901 incident on the center of the X-ray detector in the 123rd view at the position where the X-ray tube 301 faces. (The X-ray 900 and the X-ray 901 are shown with their positions shifted for convenience of drawing). That is, the X-ray 900 and the X-ray 901 pass through the same region on the subject P, although the irradiation times are different.

  By the way, when there is a moving region on the trajectory through which the X-ray 900 and the X-ray 901 are transmitted, since the irradiation times of the X-ray 900 and the X-ray 901 are different, the values of the facing data are different. Further, when the amount of movement in the moving region is large, the change in the amount of transmission is large, so that the difference in the value of the opposing data also becomes large. In order to generate the weight map 820, the reconstruction processing unit 103 calculates the absolute value of the difference between the opposing data. When scanning for two rotations is performed, 900 absolute values of the difference between the opposing data are obtained for the X-ray detection signals for 1800 views. The reconstruction processing unit 103 performs reconstruction based on the 900 absolute values calculated. This reconstruction is the same as the process of reconstructing the reference images 810 and 811 except that the projected value is not the X-ray transmission amount but the absolute value of the difference between the opposing X-ray detection signals. Since the absolute value of the difference is a value that reflects the amount of movement of the moving region, the reconstruction processing unit 103 performs reconstruction and maps the absolute value of the difference to each coordinate of the weight map 820, thereby The distribution can be reflected on the weight map 820. In addition, although the case where the scan for 2 rotations was performed was demonstrated here as an example, the scan for 1146 views may be performed, for example, and reconstruction may be performed based on 573 absolute values. It does not matter if the scan is performed.

  According to the above processing, it is possible to reduce the number of reconstruction processes by one as compared with the method of reconstructing the reference images 810 and 811 one by one and generating the weight map 820. Therefore, it is possible to reduce the processing load on the reconstruction processing unit 103 and perform higher-speed processing.

  In addition, the structure of each embodiment is not limited to what was mentioned here. For example, in each embodiment, it has been described that the reconstruction processing unit 103 reconstructs the reference images 810 and 811 and generates the weight map 820. Although the reference images 810 and 811 and the weight map 820 are shown as images in FIGS. 6, 7, and 8 for the sake of simplicity of explanation, they need not be reconstructed as image data. For example, the reconstruction processing unit 103 may generate the reconstruction of the reference images 810 and 811 and the weight map 820 as data in which numerical values are arranged.

  In each embodiment, it has been described that the X-ray detector 302 includes a plurality of rows of X-ray detection elements, but the configuration of the X-ray CT apparatus 1 is not limited thereto. For example, even if the X-ray detector 302 is composed of a single row of X-ray detection elements, reconstruction can be performed in consideration of the moving region by the method described in each embodiment.

Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 1 X-ray CT apparatus 3 Gantry 100 Control part 102 Scan control part 103 Reconstruction process part 104 Image processing part 106 Display part 107 Storage part 108 Input part 201 X-ray tube control part 202 Stand drive control part 203 Bed drive control part 300 Rotation Body 301 X-ray tube 302 X-ray detector 401 First collimator 430 Second collimator 500 Top plate 501 Top plate driver 800 Raw data 810 Reference image 811 Reference image 820 Weight map 830 Weighting function

Claims (7)

An X-ray tube that emits X-rays;
An X-ray detector that is disposed opposite to the X-ray tube and detects the X-ray and outputs an X-ray detection signal;
Reconstruction means for reconstructing a CT image based on the X-ray detection signal;
A first CT image reconstructed based on the X-ray detection signal detected in the first time domain, and a first CT image reconstructed based on the X-ray detection signal detected in the second time domain. Weighting means for setting a weighting function associated with coordinate information in the CT image based on the difference between the two CT images,
The X-ray CT apparatus, wherein the reconstruction unit reconstructs a third CT image based on the X-ray detection signal weighted according to the weighting function. A storage unit that stores at least three weighting functions;
The X-ray CT apparatus according to claim 1, wherein the weighting unit reads the weighting function from the storage unit based on the difference and sets the weighting function. The weighting function is
Based on the value of the difference, among the X-ray detection signals used for reconstruction of the third CT image, a weight for the first X-ray detection signal detected at a predetermined time is increased, and the first The X-ray CT apparatus according to claim 1, wherein the X-ray CT apparatus is set so as to reduce a weight for the second X-ray detection signal passing through an X-ray trajectory in the X-ray detection signal. The weighting means is
A weighting map that associates coordinate information in the CT image with the difference is generated based on the first CT image and the second CT image, and the weighting function is set based on the weight map. Item 4. The X-ray CT apparatus according to any one of Items 1 to 3. An X-ray tube that emits X-rays;
An X-ray detector that is disposed opposite to the X-ray tube and detects the X-ray and outputs an X-ray detection signal;
Reconstruction means for reconstructing a CT image based on the X-ray detection signal;
A first X-ray detection signal detected at a first time, and a second X-ray detection signal detected at a second time that passes through an X-ray trajectory in the first X-ray detection signal; Weighting means for setting a weighting function associated with coordinate information in the CT image based on the difference of
The X-ray CT apparatus, wherein the reconstruction unit reconstructs a third CT image based on the X-ray detection signal weighted according to the weighting function. An X-ray tube that emits X-rays;
An X-ray detector that is disposed opposite to the X-ray tube and detects the X-ray and outputs an X-ray detection signal;
In an X-ray CT apparatus comprising reconstruction means for reconstructing a CT image based on the X-ray detection signal,
A first CT image reconstructed based on the X-ray detection signal detected in the first time domain, and a first CT image reconstructed based on the X-ray detection signal detected in the second time domain. A time resolution changing means for setting a time resolution function associated with coordinate information in the CT image based on the difference between the two CT images,
The X-ray CT apparatus, wherein the reconstruction means reconstructs a CT image based on the X-ray detection signal in which the time resolution function is set. Reconstructing a first CT image based on an X-ray detection signal detected in a first time domain;
Reconstructing the second CT image based on the X-ray detection signal detected in the second time domain;
Calculating a difference between the first CT image and the second CT image, and setting a weighting function associated with coordinate information in the CT image based on the difference;
Weighting the X-ray detection signal according to the set weighting function;
A signal processing program for an X-ray CT apparatus, comprising: reconstructing a third CT image based on the weighted X-ray detection signal.

Images