JP5608441B2 - Radiation imaging apparatus and method, and program - Google Patents

Description

  The present invention relates to a radiography apparatus, method, and program for performing tomosynthesis imaging for generating a tomographic image.

  In recent years, in an X-ray imaging apparatus, in order to observe the affected area in more detail, an X-ray tube is moved to irradiate a subject with X-rays from different angles, and the acquired images are added to obtain a desired Tomosynthesis imaging capable of obtaining an image in which a tomographic plane is emphasized has been proposed. In tomosynthesis imaging, depending on the characteristics of the imaging device and the required tomographic image, the X-ray tube is moved in parallel with the X-ray detector, or moved in a circle or ellipse arc, and the subject is exposed at different irradiation angles. A plurality of captured images are acquired, and a tomographic image is generated by reconstructing these captured images using a back projection method such as a simple back projection method or a filtered back projection method. Here, in the simple backprojection method, each captured image is backprojected and backprojected onto a path according to the geometrical arrangement of the imaging system at the time of X-ray irradiation on the tomographic plane where reconstruction of the subject is desired. This is a technique for reconstructing a tomographic image of a desired cross section by adding back projection images. The filter backprojection method performs filter correction that emphasizes high frequency on a captured image to prevent blurring of the backprojected image, and backprojects and adds the filter-corrected captured image to obtain a desired cross section. This is a method for reconstructing a tomographic image of the image.

  When such tomosynthesis imaging is performed, when reconstructing a plurality of captured images acquired by imaging, it is necessary to align the captured images. For this reason, at the time of tomosynthesis imaging, a marker is attached to the subject or an imaging platform on which the subject is placed, and a plurality of captured images including the marker image are acquired by capturing the marker together with the subject. By photographing the marker together with the subject in this way, it is possible to reconstruct a tomographic image while aligning the plurality of photographed images with reference to the marker image included in each of the plurality of photographed images.

  On the other hand, the position of the X-ray source and the position of the X-ray detector are determined for each of a plurality of captured images based on a plurality of marker images included in each of a plurality of captured images obtained by performing imaging using a plurality of markers. A method for performing reconstruction based on the calculated X-ray source position and X-ray detector position has also been proposed (see Patent Documents 1 and 2).

  At the time of tomosynthesis imaging, X-rays are irradiated from various angles to an X-ray detector that projects X-rays transmitted through a subject. In the case of irradiation, the distance from the X-ray tube differs depending on the position in the region where the X-ray is irradiated on the X-ray detector. Here, since the density of the captured image is inversely proportional to the X-ray dose received by the X-ray detector, the X-ray dose that arrives changes depending on the position in the region irradiated with the X-ray, and the Cause density unevenness. For this reason, a method has been proposed in which the density of a captured image is corrected and reconstructed based on the density distribution of the captured image and the distance and angular relationship between the X-ray tube and the X-ray detector (see Patent Document 3). ).

  In addition, a method has been proposed in which the computation time is shortened by reconstructing after correcting the data representing the captured image (projection data) with a filter during reconstruction using the filtered back projection method (patent) Reference 4).

JP 2005-21345 A JP 2005-21675 A JP 2009-11639 A JP 2007-117740 A

  By the way, at the time of tomosynthesis imaging, a plurality of captured images are acquired by irradiating a subject with radiation from an X-ray tube at each of a plurality of radiation source positions. Multiple projection images are back-projected if the angle that defines the moving range of the X-ray tube (the tomographic angle) is not at a position that equally divides, that is, if the angle interval of each radiation source position with respect to the reference position is not equal When the back projection is performed by a method or the like, the balance of the information amount of each captured image in the tomographic image is not uniform. Specifically, when the source positions are set so as to equally divide the movement range of the X-ray tube, the X-ray detector is compared with the case where the respective source positions are set at equal angular intervals. The angle interval becomes smaller as X-rays are irradiated from an oblique direction. Therefore, when a tomographic image is reconstructed, an X-ray detector is irradiated with X-rays from an oblique direction. The amount of information is dense. If the information amount of the captured image in the tomographic image varies depending on the position of the X-ray source from which the captured image is acquired in this way, even if the relationship between each captured image and the source position is accurate, the tomographic image The balance of the amount of information that contributes to the angle becomes uneven, resulting in artifacts. Therefore, it is preferable that each radiation source position has an equiangular interval with respect to the reference position.

  However, depending on the imaging system, there are many cases in which the control for setting each radiation source position at an equiangular interval is complicated. For example, when the X-ray tube is moved along a linear trajectory, the moving speed of the X-ray tube is controlled non-linearly or the imaging interval is controlled non-linearly in order to make the radiation source positions equiangularly spaced. There is a need. Even if the control is performed so that the angular intervals are equal, the source positions may not actually be equal angular intervals due to mechanical misalignment of the apparatus. In particular, when the interval between the adjacent radiation source positions becomes large, the information amount of the captured image in the tomographic image becomes particularly coarse at that portion, and the artifact becomes large.

  Here, the methods described in Patent Documents 1 and 2 calculate the position of the X-ray source and the position of the X-ray detector for each of a plurality of captured images, and calculate the position of the X-ray source and the X-ray detector. In this method, reconstruction is performed based on the positions of the positions, and the angular interval of the source positions with respect to the reference position is not a problem. In addition, the method described in Patent Document 3 corrects density unevenness due to a difference in distance from the X-ray tube in one captured image, and uses the reference position as a reference as in Patent Documents 1 and 2. It does not matter the angular spacing of the source positions. In Patent Document 4, reconstruction is performed after correction using a filter on data representing a photographed image at the time of reconstruction using the filter back projection method. Similarly, a radiation source based on a reference position is used. It does not matter the angular spacing of the positions.

  The present invention has been made in view of the above circumstances, and an object thereof is to enable easy acquisition of a tomographic image with reduced artifacts using a captured image acquired by tomosynthesis imaging.

A radiation imaging apparatus according to the present invention includes a radiation source for irradiating a subject with radiation,
Detecting means for detecting radiation transmitted through the subject;
Moving the radiation source relative to the detection means within a moving range calculated based on a predetermined tomographic angle with respect to a predetermined reference point on a reference plane defining a range for acquiring a tomographic image; Image acquisition means for irradiating the subject with the radiation at a plurality of radiation source positions due to movement of the radiation source and acquiring a plurality of captured images corresponding to the plurality of radiation source positions;
Interpolation means for generating an interpolated captured image for interpolating the captured image acquired at the adjacent source position when an interval between adjacent source positions is equal to or greater than a predetermined threshold;
Image reconstructing means for generating a tomographic image of the subject by back-projecting the plurality of captured images and the interpolated captured image onto a desired cross-section of the subject and adding the back-projected captured images; It is characterized by that.

  “Move the radiation source relative to the detection means” means both when the detection means is fixed and only the radiation source is moved, and when both the detection means and the radiation source are moved in synchronization. including.

  Examples of the “reference plane that determines the range for acquiring the tomographic image” include the top surface of the imaging table on which the subject is placed, the detection surface of the detection means, and the surface closest to the detection means for the region of interest when the region of interest is set. Alternatively, any tomographic plane or the like in the subject can be used.

  The “tomographic angle” is an angle that faces two end portions that define a moving range of the radiation source from a reference point on the reference plane.

  The radiation imaging apparatus according to the present invention may further include a source position acquisition unit that acquires information on the plurality of source positions when the captured image is acquired.

  In the radiographic apparatus according to the present invention, the interval between the adjacent radiation source positions may be an angular interval based on the reference point of the radiation source position from which the adjacent captured image is acquired.

  Further, in the radiographic apparatus according to the present invention, the image reconstructing means is configured to determine an angle of the interpolated source position corresponding to the interpolated captured image and the source position from which each captured image is acquired with reference to the reference point. The larger the interval, the higher the weight of each of the back-projected captured image and the interpolated captured image, and the weighted addition may be performed to generate a tomographic image of the subject.

  “An angle with respect to the reference point” is an angle formed by an axis extending in the moving direction of the radiation source parallel to the reference plane passing through the reference point and a straight line connecting the reference point and the source position. Can be calculated as a difference value of angles with reference to a reference point for two source positions adjacent to a certain source position.

  The radiographic apparatus according to the present invention may further include angle interval acquisition means for acquiring the angle interval.

A radiation imaging method according to the present invention includes a radiation source for irradiating a subject with radiation,
Detecting means for detecting radiation transmitted through the subject;
Moving the radiation source relative to the detection means within a moving range calculated based on a predetermined tomographic angle with respect to a predetermined reference point on a reference plane defining a range for acquiring a tomographic image; A radiation imaging apparatus comprising: an image acquisition unit configured to irradiate the subject with the radiation at a plurality of radiation source positions due to movement of the radiation source and to acquire a plurality of captured images corresponding to the plurality of radiation source positions, respectively. A radiography method in
When an interval between adjacent source positions is equal to or greater than a predetermined threshold value, an interpolated captured image that interpolates captured images acquired at the adjacent source positions is generated.
The plurality of captured images and the interpolated captured image are back-projected on a desired cross section of the subject, and the tomographic image of the subject is generated by adding the back-projected captured images. is there.

  In addition, you may provide as a program for making a computer perform the radiography method by this invention.

  As described above, if the angular intervals of the respective source positions with reference to the reference point are not equal intervals, the balance of the information amount of the captured images is not uniform in the tomographic images when a plurality of captured images are back-projected, and the angular intervals The amount of information acquired from a captured image obtained at a position with a small source becomes denser, and the amount of information of the captured image becomes denser in the tomographic image. When the information amount of the captured image in the tomographic image is thus dense, even if the relationship between each captured image and the source position is accurate, the balance of the information amount contributing to the tomographic image by angle is uneven. And artifacts occur. In this case, it is preferable to set each radiation source position at equal angular intervals with reference to the reference point. However, in order to make the radiation source positions at equal angular intervals, control when moving the radiation source becomes complicated. . In particular, when the interval between adjacent radiation source positions becomes large, the information amount of the captured image in the tomographic image becomes particularly coarse at that portion, and the artifact becomes large.

  According to the present invention, when an interval between adjacent source positions is equal to or greater than a predetermined threshold value, an interpolated captured image that interpolates captured images acquired at adjacent source positions is generated, and the subject is desired. A tomographic image of a subject is generated by back-projecting a plurality of captured images and interpolated captured images on a cross section and adding the back-projected captured images. For this reason, when the angular interval between adjacent source positions becomes large, the interpolated captured image is interpolated between the captured images at the source position, so that the information amount of the captured image in the tomographic image becomes particularly coarse. Therefore, a high-quality tomographic image with few artifacts can be generated.

  Also, the greater the angular interval with respect to the reference point of the interpolated source position corresponding to the interpolated captured image and the source position from which each captured image was acquired, the greater the weighting of each backprojected captured image and interpolated captured image By increasing the weight and adding weights, even if the angular intervals of multiple radiation source positions are not uniform, the balance of the amount of information that contributes to the tomographic image is equalized, thereby reducing the artifacts. An image quality tomographic image can be generated.

  In addition, by acquiring information on a plurality of source positions when a captured image is acquired, an interpolated captured image can be generated or weighted according to the actual source position at the time of capturing. In addition, it is possible to correct the deviation of the backprojection path caused by the positional deviation at the time of photographing, and at the same time, to correct the image quality fluctuation caused by the density of the information amount by angle, thereby producing a high-quality tomographic image with less artifacts. Can be generated.

Schematic of an X-ray imaging apparatus to which the radiation imaging apparatus according to the first embodiment of the present invention is applied. Diagram for explaining calculation of movement range of X-ray tube The top view of the top plate of the imaging stand explaining the marker used in 1st Embodiment, and the arrangement position of a marker The figure which shows the process which a reconstruction part performs roughly The figure for demonstrating calculation of the weighting coefficient in 1st Embodiment. The figure for demonstrating calculation of the interpolation picked-up image and update of a weighting coefficient in 1st Embodiment. The flowchart which shows the process performed in 1st Embodiment. Diagram for explaining the density of information in tomographic images Schematic of an X-ray imaging apparatus to which a radiation imaging apparatus according to the second embodiment of the present invention is applied. The flowchart which shows the process performed in 2nd Embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic view of an X-ray imaging apparatus to which the radiation imaging apparatus according to the first embodiment of the present invention is applied. As shown in FIG. 1, an X-ray imaging apparatus 10 according to the first embodiment is for tomosynthesis imaging, and includes an X-ray tube 12 and a flat panel X-ray detector (hereinafter simply referred to as a detector). 14. The X-ray tube 12 moves along a straight line or an arc by the moving mechanism 16 and irradiates the subject 2 on the imaging table top 4 with X-rays at a plurality of positions on the moving path. In the present embodiment, the X-ray tube 12 is moved in the direction of arrow A along a linear trajectory. Note that the amount of X-ray irradiation to the subject 2 is controlled to be a predetermined amount by a control unit described later.

  Further, a collimator (irradiation field stop) 6 is connected to the X-ray tube 12 so that the operator can set an X-ray range (irradiation range) irradiated to the subject 2. When setting the irradiation range using the collimator 6, visible light is irradiated to the subject 2 through the collimator 6 instead of X-rays. The visible light is emitted from an irradiation field lamp (not shown) provided in the collimator 6. Thus, the operator can set the X-ray irradiation range by adjusting the range of visible light irradiated on the subject 2 using the collimator 6. In this embodiment, a marker is arranged on the imaging table top 4 as will be described later, and imaging is performed so that the marker is included in the plurality of captured images together with the subject 2.

  The detector 14 is disposed so as to face the X-ray tube 12 with the imaging table top plate 4 on which the subject 2 is placed interposed therebetween in order to detect X-rays transmitted through the subject 2. The detector 14 is moved along a straight line or an arc as necessary by the moving mechanism 18 and detects X-rays transmitted through the subject 2 at a plurality of positions on the moving path. In the present embodiment, the detector 14 is moved in the direction of arrow B along a linear trajectory.

  The X-ray imaging apparatus 10 includes an image acquisition unit 20 and a reconstruction unit 22. The image acquisition unit 20 moves the X-ray tube 12 along a straight line, irradiates the subject 2 with X-rays at a plurality of source positions by the movement of the X-ray tube 12, and detects X-rays transmitted through the subject 2 as a detector. 14 to obtain a plurality of captured images at a plurality of moving source positions. The reconstruction unit 22 reconstructs a plurality of captured images acquired by the image acquisition unit 20 to generate a tomographic image indicating a desired cross section of the subject 2. In the present embodiment, the reconstruction unit 22 reconstructs these captured images using a back projection method such as a simple back projection method or a filtered back projection method, and generates a tomographic image. In this embodiment, the filter back projection method is used.

  The X-ray imaging apparatus 10 includes an operation unit 24, a display unit 26, and a storage unit 28. The operation unit 24 includes a keyboard, a mouse, or a touch panel type input device, and receives an operation of the X-ray imaging apparatus 10 by an operator. It also accepts input of various information such as imaging conditions and information correction instructions necessary for performing tomosynthesis imaging. In the present embodiment, each unit of the X-ray imaging apparatus 10 operates in accordance with information input from the operation unit 24 by the operator. The display unit 26 is a display device such as a liquid crystal monitor, and displays a message necessary for the operation in addition to the captured image acquired by the image acquisition unit 20 and the tomographic image reconstructed by the reconstruction unit 22. The display unit 26 may include a speaker that outputs sound. The storage unit 28 stores various parameters for setting imaging conditions necessary for operating the X-ray imaging apparatus 10. Various parameters are stored in the storage unit 28 as standard values corresponding to the imaging region, and are corrected by an operator giving instructions from the operation unit 24 as necessary.

  Parameters for setting imaging conditions include a reference plane, a tomographic angle, a source distance, a shot number, a shot interval, a tube voltage and a tube current of the X-ray tube 12, an X-ray exposure time, and the like. Of these parameters, the number of shots, the shot interval, the tube voltage and tube current of the X-ray tube 12, and the X-ray exposure time can be used as imaging conditions as they are.

  FIG. 2 is a diagram for explaining various parameters. The reference plane is a plane that defines a range in which tomographic images are acquired. For example, the top plate surface of the imaging table top plate 4, the detection surface of the detector 14, or an arbitrary tomographic plane in the subject 2 can be used. In FIG. 2, a plane (hereinafter referred to as a center plane) that bisects the thickness of the subject 2 is used as a reference plane. The tomographic angle is an angle that faces two end portions that define the movement range of the X-ray tube 12 from the reference point B0 on the reference plane. Here, since the detection surface of the detector 14 and the movement path of the X-ray tube 12 are parallel, the distance closest to the detection surface of the detector 14 on the movement path of the X-ray tube 12 is the radiation source distance. To do. 2 and the following description, the direction parallel to the movement path of the X-ray tube 12 is the x direction, the direction perpendicular to the movement path of the X-ray tube 12 is the z direction, and the direction perpendicular to the paper surface is the y direction. To do.

  The number of shots is the number of times of imaging while the X-ray tube 12 moves from end to end within the range of the tomographic angle. The shot interval is a time interval between shots.

  In FIG. 2 and the following description, the movement range of the X-ray tube 12 is s0, the distance between the X-ray tube 12 and the detection surface of the detector 14 (ie, the source distance) is sz, the tomographic angle is θ, and the detection is performed. The distance between the detection surface of the device 14 and the reference surface (that is, the center surface of the subject 2) is d0, and the distance between the detection surface of the detector 14 and the top surface of the imaging table top 4 is mz. In addition, as a predetermined reference point B0 on the reference plane, an intersection of a perpendicular passing through the center of gravity of the detector 14 and the reference plane is used.

  The X-ray imaging apparatus 10 includes a calculation unit 30. The calculation unit 30 calculates imaging conditions such as the movement range of the X-ray tube 12 according to the parameters stored in the storage unit 28.

  Here, referring to the relationship shown in FIG. 2, the movement range s0 of the X-ray tube 12 can be calculated from the source distance sz, the distance d0, and the tomographic angle θ. That is, if the intersection of the perpendicular passing through the reference point B0 and the movement path of the X-ray tube 12 is the origin O1, the distance between the reference plane and the X-ray tube 12 is sz-d0. The movement range s0 of the tube 12 is calculated as-(sz-d0) · tan (θ / 2) to (sz-d0) · tan (θ / 2). This determines the positions of both ends of the calculated movement range s0.

  Further, the calculation unit 30 equally divides the moving range s0 of the X-ray tube 12 by the number of shots, thereby calculating the position of the X-ray tube 12 in each radiographing (hereinafter referred to as a radiation source position). Thereby, the source positions S1 to Sn of the X-ray tube 12 can be calculated so as to equally divide the movement range s0.

  Moreover, the calculating part 30 calculates imaging | photography time and a radiation source travel speed as imaging | photography conditions. The shooting time can be calculated by the number of shots × shot interval. The radiation source traveling speed can be calculated from the moving range s0 / imaging time.

  In addition, the X-ray imaging apparatus 10 includes a source position calculation unit 32 that calculates the source positions S1 to Sn of the X-ray tube 12 using a marker image included in the captured image. FIG. 3 is a plan view of a marker used in the present embodiment and the top plate of the photographing table for explaining the arrangement position of the marker. As shown in FIG. 3, in the present embodiment, four circular markers M1 to M4 are used, and the operator 2 can overlap the subject 2 and the markers M1 to M4 in the X-ray irradiation direction. Four markers M <b> 1 to M <b> 4 are arranged on the imaging stand top plate 4. The markers M1 to M4 are made of a material such as lead having a high X-ray absorption rate. The size of the markers M1 to M4 is about 1 cm, and each has a hole with a unique shape. Thereby, each of the four markers M1 to M4 included in the captured image can be identified. In FIG. 3, the sizes of the markers M1 to M4 are enlarged. Further, the shape of the marker is not limited to a circle, and any known shape can be used. Also, the number of markers is not limited to four, and may be any number greater than or equal to one.

  The radiation source position calculation unit 32 detects marker images of the markers M1 to M4 from each of a plurality of captured images acquired by tomosynthesis imaging, and uses the detected marker image position (center position of the marker image) as a radiation source. Calculate the position. Here, a certain relationship is established among the positions of the X-ray tube 12, the detector 14, the markers M1 to M4, and the marker image. The X-ray tube 12, the markers M1 to M4, and the marker image are all point positions, and the position of the detector 14 is a plane position. The positions of the markers M1 to M4 are acquired when the markers M1 to M4 are arranged on the imaging table top plate 4. The radiation source position calculation unit 32 calculates a relational expression related to the relationship between the positions of the markers M1 to M4, the marker image, the X-ray tube 12 and the detector 14 by using, for example, the method described in Patent Document 2. An equation is obtained by substituting information on the positions of the markers M1 to M4 and the position of the marker image into the relational expression. By decomposing this equation for each coordinate component (x, y, z) of the X-ray imaging apparatus, three equations are obtained for each marker. Then, the calculated position of the X-ray tube 12 (that is, the position of the source) and the position of the detector 14 are set as unknown information, and by solving twelve simultaneous equations derived from the four markers M1 to M4, the position of the source is determined. The positions of S1 to Sn and the detector 14 are calculated. When the detector 14 is not moved, the position of the detector 14 is fixed, so that the source position can be calculated by a simpler equation.

  The radiation source position calculation unit 32 calculates the radiation source position based on information on the positions of the markers M1 to M4 and the position of the marker image using another known method such as the above-mentioned Patent Document 1. May be. Here, when the technique described in Patent Document 1 is used, the radiation source position calculation unit 32 acquires information on the positions of the marker images of the markers M1 to M4 from each of the plurality of captured images, and the markers M1 to M4. The projection positions of the markers M1 to M4 on the detector 14 are estimated by performing a projection calculation on the assumption that the position of is known. For the estimation of the projection position, a parameter for determining the source position, a parameter for the position of the origin of the detector 14, and a vector for determining the orientation of the detection surface are used. Accordingly, the source position parameter, the origin position parameter of the detector 14, and the vector for determining the orientation of the detection surface are determined so that the error between the estimated projection position and the detected marker image position is minimized. By doing so, the information of the radiation source position can be calculated.

  In addition, the X-ray imaging apparatus 10 includes a weight calculation unit 34 that calculates a weight coefficient used when reconstructing a plurality of captured images. Here, in the present embodiment, the reconstruction unit 22 reconstructs a tomographic image using a filtered back projection method. FIG. 4 is a diagram schematically showing processing performed by the reconstruction unit 22. In the present embodiment, a plurality of photographed images G1 to Gn are acquired by photographing the subject 2 at a plurality of radiation source positions S1 to Sn. Then, filter corrections F1 to Fn for emphasizing a high frequency by a one-dimensional filter are performed on each of the plurality of photographed images G1 to Gn, and the photographed images G1 ′ to Gn ′ having been subjected to the filter correction are reconstructed on the subject 2. On the desired tomographic plane, a tomographic image D0 having a desired cross section is reconstructed by back projecting and adding to a path determined by a plurality of radiation source positions.

  At this time, in the present embodiment, the greater the angular interval with respect to the reference point B0 of the source positions S1 to Sn from which the captured images G1 to Gn are acquired, the greater the weighting, and the captured image subjected to filter correction. G1 ′ to Gn ′ are added. The weight calculation unit 34 calculates weight coefficients W1 to Wn when adding the captured images G1 ′ to Gn ′ that have been subjected to the filter correction. The weight calculation unit 34 corresponds to the angle interval acquisition unit of the present invention.

  FIG. 5 is a diagram for explaining calculation of the weighting coefficient. As shown in FIG. 5, the weight calculation unit 34 converts the source positions S1 to Sn calculated by the source position calculation unit 32 into angles θ1 to θn with the reference point B0 as a reference. Specifically, the angles θ1 to θn are angles formed by an axis in the x direction parallel to the reference plane passing through the reference point B0 and a straight line connecting the reference point B0 and the radiation source positions S1 to Sn. Here, since the positions of the radiation source positions S1 to Sn are calculated by the radiation source position calculation unit 32 and the reference point B0 is known, the angles θ1 to θn can be calculated geometrically.

  Next, the weight calculation unit 34 calculates the angle interval dθi (i = 1 to n) between the radiation source positions S1 to Sn by the following equation (1).

dθi = (θ _{i + 1} −θ _i-1 ) / 2 (1)
That is, for the target source position Si, a value that equally divides the angles θi + 1 and θi−1 of the adjacent source positions Si + 1 and Si−1 is calculated as the angular interval of the target source position Si. . The angle spacing d1, dn source position S1, Sn of both ends are calculated by the respective _{d2-d1, dn-d n} -1. The weight calculation unit 34 outputs the angle interval dθi calculated in this way as a weight coefficient Wi for the captured images G1 ′ to Gn ′ subjected to filter correction. Note that a value obtained by normalizing the angular interval dθi according to the following equation (2) may be used as the weighting factor Wi.

Wi = dθi / (Σdθi) (2)
The X-ray imaging unit 10 includes an interpolation unit 36. The interpolation unit 36 calculates a difference Δθi between the angles θi and θi + 1 of the adjacent radiation source positions Si and Si + 1, and determines whether or not the difference Δθi is equal to or greater than a predetermined threshold value Th1. When the difference Δθi is equal to or larger than the threshold value Th1, the interpolated captured image Ghi is generated by interpolating the captured images Gi and Gi + 1 acquired at the source positions Si and Si + 1. The interpolated captured image Ghi may be obtained by calculating a simple average of pixels corresponding to the captured images Gi and Gi + 1. Here, in FIG. 5, if the difference Δθ2 between the angle θ2 of the radiation source position S2 and the angle θ3 of the radiation source position S3 is equal to or greater than the threshold value Th1, the interpolation unit 36 determines that the captured image G2 Is averaged to generate an interpolated photographed image.

  The interpolated captured image is used in place of the captured image acquired when the source position is at an angle that equally divides the difference between the angles θi and θi + 1 of the adjacent source positions Si and Si + 1. Therefore, when the interpolated captured image is calculated, the weight calculating unit 34 virtually sets the source position corresponding to the interpolated captured image, updates the source position, and sets the updated source positions S1 to Sn. The weighting factor Wi is updated by recalculating the weighting factor Wi by converting the angle θ1 to θn with the reference point B0 as a reference. For example, when a virtual source position is set between the source position S2 and the source position S3 in FIG. 5, the weight calculation unit 34 sets the virtual source position S3 as shown in FIG. Then, the source positions S3, S4... Are updated to the source positions S4, S5. Also, the angles θ3, θ4,... Corresponding to the radiation source positions S3, S4,. Then, the weighting factor Wi is updated by recalculating the weighting factor Wi based on the updated angle.

  Further, the X-ray imaging apparatus 10 includes a control unit 38 for controlling each unit of the X-ray imaging apparatus 10. The control unit 38 controls each unit of the X-ray imaging apparatus 10 according to an instruction from the operation unit 24. Further, the control unit 38 determines the X-ray irradiation amount to the subject 2 based on the X-ray tube 12 based on the tube voltage and tube current of the X-ray tube 12 and the X-ray exposure time stored in the storage unit 28. Control.

  Next, processing performed in the first embodiment will be described. FIG. 7 is a flowchart showing processing performed in the first embodiment. In the following description, it is assumed that only the X-ray tube 12 is moved and the detector 14 is not moved to perform tomosynthesis imaging. When the operation unit 24 receives an instruction to start processing by the operator, the control unit 38 starts processing, performs tomosynthesis imaging while moving the X-ray tube 12 (step ST1), and the image acquisition unit 20 performs a plurality of captured images. Is acquired (step ST2). Next, the radiation source position calculation unit 32 calculates the radiation source position based on the positions of the markers M1 to M4 and the position of the marker image (step ST3). Then, the weight calculation unit 34 calculates the weight coefficient Wi for the plurality of captured images based on the calculated radiation source position (step ST4).

  On the other hand, the interpolation unit 36 calculates the angle difference between adjacent source positions for the plurality of source positions calculated by the source position calculation unit 32, and the source position at which the angle difference is equal to or greater than the threshold Th1. Is determined (step ST5). If step ST5 is positive, the interpolation unit 36 generates an interpolated photographed image (step ST6), and the weight calculation unit 34 updates the weight coefficient Wi (step ST7). Subsequent to step ST7 and when step ST5 is negative, the reconstruction unit 22 reconstructs a plurality of captured images including an interpolated captured image by a filter back projection method using the weighting factor Wi, and generates a tomographic image. Generate (step ST8), and the process ends. That is, filter corrections F1 to Fn for enhancing high frequency by a one-dimensional filter are performed on each of the plurality of captured images G1 to Gn including the interpolated captured image, and the captured images G1 ′ to Gn ′ that have been subjected to the filter correction are obtained as subject. On the tomographic plane for which reconstruction of 2 is desired, the tomographic image is back-projected onto a path determined by a plurality of radiation source positions, and the back-projected captured images G1 ′ to Gn ′ are weighted and added by the weighting factor Wi. Reconstruct the image. Note that the generated tomographic image is stored in a storage device such as an HDD (not shown) or transmitted to an external server via a network.

  Here, if the angular intervals dθi of the radiation source positions S1 to Sn with reference to the reference point B0 are not equal, the information amount of the captured images G1 to Gn in the tomographic image when the plurality of captured images G1 to Gn are back-projected. And the amount of information becomes denser in the tomographic image, and the amount of information in the captured images G1 to Gn becomes denser in the tomographic image. FIG. 8 is a diagram for explaining the density of information in a tomographic image. FIG. 8 is a diagram in which each of a plurality of captured images is subjected to Fourier transform, and the information of the captured images is represented in Fourier space. In FIG. 8, the information of the captured image when the radiation source position is at the origin O1 of the movement range s0 is located on the fx axis, and the origin in the Fourier space is centered as the radiation source position is farther from the origin O1. Thus, the information of the photographed image is expressed in an inclined manner.

  Here, FIG. 8A shows information of each captured image when the angular intervals of the radiation source positions S1 to Sn are equal. As shown in FIG. 8A, when the angular intervals of the radiation source positions S1 to Sn are equal, the information of the captured image is arranged at equal angular intervals in the Fourier space. On the other hand, when the angular interval between adjacent radiation source positions is increased, the angular interval of information of the captured image is increased as shown in FIG. Here, adding back-projected captured images is equivalent to integrating information in Fourier space. For this reason, as shown in FIG. 8A, when the angular intervals of the radiation source positions S1 to Sn are equal, the information amount of each captured image does not vary in the tomographic image, but FIG. As shown in FIG. 5, when there is a place where the angular interval between adjacent radiation source positions is large, the information amount of each captured image is coarsely and densely generated in the tomographic image. Further, as shown in FIG. 8C, if the angular intervals of the radiation source positions S1 to Sn are not equal, the information amount of each captured image in the tomographic image is coarse. Thereby, the balance of the information amount of each captured image is not uniform in the tomographic image, and the amount of information included in the captured image increases as the angle interval is smaller. When the information amount of the captured image in the tomographic image is thus dense, even if the relationship between the captured images G1 to Gn and the source positions S1 to Sn is accurate, the information amount that contributes to the tomographic image is classified according to the angle. The balance becomes uneven and artifacts occur. In this case, it is preferable that the radiation source positions S1 to Sn be equiangular intervals with reference to the reference point B0. However, in order to make the radiation source positions equiangular intervals, the X-ray tube 12 is moved. The control becomes complicated.

  According to the present embodiment, the difference Δθi between the angles θi and θi + 1 of the radiation source positions Si and Si + 1 adjacent to each other is calculated, and when the difference Δθi is equal to or greater than the threshold Th1, the imaging acquired at the adjacent radiation source position. An interpolated photographed image for interpolating an image is generated. For this reason, there is no portion where the information amount of the captured image in the tomographic image becomes particularly coarse. Furthermore, in the present embodiment, the greater the angular interval dθi with respect to the reference point B0 of the source positions S1 to Sn from which the captured images G1 to Gn including the interpolated captured images are acquired, the weighting is increased and the inverse is performed. The projected captured images G1 to Gn are weighted and added. For this reason, even if the angular intervals dθi are unequal, the balance by angle of contribution of the captured images G1 to Gn to the tomographic image can be made uniform according to the density of the angular intervals dθi. Therefore, according to the present embodiment, it is possible to correct the balance of the amount of information contributing to the tomographic image for each angle, and as a result, it is possible to generate a high-quality tomographic image with few artifacts.

  In the first embodiment, when the weight calculation unit 34 calculates the weight coefficient Wi and the interpolation unit 36 generates the interpolated photographed image, the source position is virtually set and the weight coefficient Wi is set. Although updated, the weighting factor Wi is not calculated first, and after generating the interpolated captured image, the weighting factor Wi is calculated using the source position of the captured image and the virtual source position of the interpolated captured image. You may make it do.

  Next, a second embodiment of the present invention will be described. FIG. 9 is a schematic view of an X-ray imaging apparatus to which the radiation imaging apparatus according to the second embodiment of the present invention is applied. In the second embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed description thereof is omitted here. The X-ray imaging apparatus 10A according to the second embodiment omits the weight calculation unit 34 for calculating the weighting coefficient Wi, and the reconstruction unit 22 simply adds the back-projected captured images to reconstruct a tomographic image. This is different from the first embodiment.

  Next, processing performed in the second embodiment will be described. FIG. 10 is a flowchart showing processing performed in the second embodiment. In the following description, it is assumed that only the X-ray tube 12 is moved and the detector 14 is not moved to perform tomosynthesis imaging. When the operation unit 24 receives an instruction to start processing by the operator, the control unit 38 starts processing, performs tomosynthesis imaging while moving the X-ray tube 12 (step ST11), and the image acquisition unit 20 performs a plurality of captured images. Is acquired (step ST12). Next, the radiation source position calculation unit 32 calculates the radiation source position based on the positions of the markers M1 to M4 and the position of the marker image (step ST13).

  On the other hand, the interpolation unit 36 calculates the angle difference between adjacent source positions for the plurality of source positions calculated by the source position calculation unit 32, and the source position at which the angle difference is equal to or greater than the threshold Th1. Is determined (step ST14). When step ST14 is affirmed, the interpolation unit 36 generates an interpolated captured image (step ST15), and the reconstruction unit 22 reconstructs a plurality of captured images including the interpolated captured image by the filter back projection method to obtain a tomographic image. Is generated (step ST16), and the process is terminated. That is, filter corrections F1 to Fn for enhancing high frequency by a one-dimensional filter are performed on each of the plurality of photographed images G1 to Gn, and the photographed images G1 ′ to Gn ′ having been subjected to the filter correction are reconstructed of the subject 2. On the desired tomographic plane, the tomographic image is reconstructed by back-projecting onto a path defined by a plurality of radiation source positions and simply adding the back-projected captured images G1 ′ to Gn ′. Note that the generated tomographic image is stored in a storage device such as an HDD (not shown) or transmitted to an external server via a network.

  As a result, there is no portion where the information amount of the captured image in the tomographic image is particularly coarse, so that a high-quality tomographic image with few artifacts can be generated.

  In the first and second embodiments, the tomographic image is reconstructed using the filter backprojection method. However, the tomographic image may be reconstructed using the simple backprojection method.

  In the first and second embodiments, the interpolated captured image is generated when the difference in angle between the adjacent source positions is equal to or greater than the threshold Th1, but the adjacent source positions are not detected. An interpolated photographed image may be generated when the difference in the x direction is equal to or greater than a threshold value.

  In the first and second embodiments, only the X-ray tube 12 is moved, but the X-ray tube 12 and the detector 14 may be moved in synchronization.

  In the first and second embodiments described above, the subject is placed on the imaging stand in the supine position to perform tomosynthesis imaging, but also when tomosynthesis imaging is performed using a standing imaging platform. Of course, the present invention can be applied.

  In the second embodiment, the position of the X-ray tube 12, that is, the radiation source position is calculated using the marker image. However, a sensor for detecting the radiation source position is provided, and the radiation source position detected by the sensor is provided. May be used.

2 subject 4 imaging table top plate 6 collimator 10, 10A X-ray imaging device 12 X-ray tube 14 detector 16, 18 moving mechanism 20 image acquisition unit 22 reconstruction unit 24 operation unit 26 display unit 28 storage unit 30 calculation unit 32 line Source position calculation unit 34 Weight calculation unit 36 Interpolation unit 38 Control unit

Claims (7)

A radiation source for irradiating the subject with radiation;
Detecting means for detecting radiation transmitted through the subject;
A moving range based on the reference point is calculated based on a predetermined tomographic angle facing two ends that define the moving range of the radiation source from a predetermined reference point on a reference plane that determines a range for acquiring a tomographic image. , within the movement range which is the calculated, the radiation source is moved relative to the detection means, by irradiating the radiation to the subject at a plurality of radiation source positions by the movement of said radiation source, said plurality Image acquisition means for acquiring a plurality of captured images corresponding to each of the radiation source positions;
Interpolation means for generating an interpolated captured image for interpolating the captured image acquired at the adjacent source position when an interval between adjacent source positions is equal to or greater than a predetermined threshold;
Image reconstructing means for generating a tomographic image of the subject by back-projecting the plurality of captured images and the interpolated captured image onto a desired cross-section of the subject and adding the back-projected captured images; A radiographic apparatus characterized by that.   The radiation imaging apparatus according to claim 1, further comprising a radiation source position acquisition unit configured to acquire information on the plurality of radiation source positions when the captured image is acquired.   The radiation imaging apparatus according to claim 1, wherein the interval between the adjacent radiation source positions is an angle interval with respect to the reference point of the radiation source position from which the adjacent captured image is acquired.   The image reconstructing means is configured such that the back-projected captured image becomes larger as the angular interval with respect to the reference point of the interpolation source position corresponding to the interpolated captured image and the source position from which each captured image is acquired is larger. 4. The radiographic imaging apparatus according to claim 1, wherein the radiographic imaging system is a unit that generates a tomographic image of the subject by increasing the weighting and adding the weighting for each of the interpolated captured images. 5. apparatus.   5. The radiation imaging apparatus according to claim 3, further comprising angle interval acquisition means for acquiring the angle interval. A radiation source for irradiating the subject with radiation;
Detecting means for detecting radiation transmitted through the subject;
A moving range based on the reference point is calculated based on a predetermined tomographic angle facing two ends that define the moving range of the radiation source from a predetermined reference point on a reference plane that determines a range for acquiring a tomographic image. , within the movement range which is the calculated, the radiation source is moved relative to the detection means, by irradiating the radiation to the subject at a plurality of radiation source positions by the movement of said radiation source, said plurality A radiation imaging method in a radiation imaging apparatus comprising image acquisition means for acquiring a plurality of captured images corresponding to each of the radiation source positions,
When an interval between adjacent source positions is equal to or greater than a predetermined threshold value, an interpolated captured image that interpolates captured images acquired at the adjacent source positions is generated.
Radiation imaging, wherein the tomographic image of the subject is generated by back-projecting the plurality of captured images and the interpolated captured image onto a desired cross section of the subject and adding the back-projected captured images Method. A radiation source for irradiating the subject with radiation;
Detecting means for detecting radiation transmitted through the subject;
A moving range based on the reference point is calculated based on a predetermined tomographic angle facing two ends that define the moving range of the radiation source from a predetermined reference point on a reference plane that determines a range for acquiring a tomographic image. , within the movement range which is the calculated, the radiation source is moved relative to the detection means, by irradiating the radiation to the subject at a plurality of radiation source positions by the movement of said radiation source, said plurality A program for causing a computer to execute a radiation imaging method in a radiation imaging apparatus comprising image acquisition means for acquiring a plurality of captured images corresponding to each of the radiation source positions,
A procedure for generating an interpolated captured image for interpolating a captured image acquired at the adjacent source position when an interval between adjacent source positions is equal to or greater than a predetermined threshold;
Causing the computer to execute a procedure of back-projecting the plurality of captured images and the interpolated captured image on a desired cross-section of the subject and adding the back-projected captured images to generate a tomographic image of the subject. A program characterized by that.

Images