JP5406063B2 - Reconstruction calculation device, reconstruction calculation method, and X-ray CT apparatus - Google Patents

Description

  The present invention relates to an X-ray CT image reconstruction calculation process using a successive approximation method.

  The X-ray CT apparatus irradiates X-rays from around the subject, collects data relating to the intensity of the X-rays transmitted through the subject, and collects X-rays inside the subject based on the collected data. This is an apparatus for imaging distribution information of absorption coefficients.

  Image reconstruction methods are roughly classified into an analytical reconstruction method and an algebraic reconstruction method. Among these image reconstruction methods, analytical reconstruction methods include, for example, a Fourier transform method, a filtered back projection method (hereinafter referred to as FBP method), and a superposition integration method. For example, there is a successive approximation reconstruction method represented by the MLEM (Maximum Likelihood Extraction Maximization) method and the OSEM (Ordered Subset Extraction Maximization) method. Among them, the analytical reconstruction method currently in practical use is applied to a multi-slice CT having a wide cone angle (expansion angle in the slice direction of the X-ray beam), and the reconstruction algorithm is incomplete due to the incompleteness of the reconstruction algorithm. There is a problem of generating beam artifacts. On the other hand, it is known that the algebraic reconstruction method has higher completeness than the analytical reconstruction method, but on the other hand, a recursive operation requires a long calculation time. For this reason, the algebraic reconstruction method has been conventionally used in the field of nuclear medicine images and has not spread in the field of X-ray CT images. However, the problem of the calculation time of the successive approximation reconstruction method (algebraic reconstruction method) is being solved by the recent development of computer technology, and the successive approximation reconstruction method is used for image formation of an X-ray CT apparatus. Japanese Patent Application Laid-Open No. 2004-228561 discloses that image quality is improved.

  In Patent Document 1, after the image reconstruction process using the FBP method described above is performed, the image quality improvement using the OSEM method is performed as post-processing in order to improve the image quality of the reconstructed image for the region of interest (ROI). Processing is performed. In the image quality improvement processing, a comparison image in which the reprojection image and the projection image are compared at the same angle is created, a standard deviation SD of the comparison image is calculated, and whether or not further image quality improvement is necessary depending on the magnitude of the standard deviation SD. It is determined, and the image quality improvement process is repeated until the end condition is satisfied.

JP 2006-25868 A

  However, if the image quality improvement processing using the OSEM method (sequential approximation reconstruction method) is performed after the image reconstruction processing by the FBP method as in Patent Document 1 described above, the effect of the reconstruction filter used in the FBP method is obtained. It will be damaged. For example, an image having a predetermined frequency characteristic may be required for diagnosis by a doctor or the like, but the frequency characteristic adjusted by the reconstruction filter adopted in the above FBP method is damaged by the subsequent OSEM method. .

  The present invention has been made in view of the above problems, and a reconstruction calculation device, a reconstruction calculation method, and an X-ray CT image that are effective for medical diagnosis can be created more quickly and more accurately. And an X-ray CT apparatus.

  In order to achieve the above-described object, the first invention performs a reconstruction process including a successive approximation process on X-ray projection data acquired by imaging with a predetermined scan locus, and generates a first image. Reconstructing means, forward projection data generating means for generating forward projection data by forward projecting the first image, and analytical reconstruction in which desired reconstruction conditions are set for the forward projection data A second reconstructing unit that performs processing and reconstructs a final image, and the forward projection data generating unit includes artifacts in the reconstructing process of the second reconstructing unit as compared with the photographing. The reconstruction calculation device is characterized in that the first image is projected in a forward direction with a scan trajectory with little mixing.

  According to a second aspect of the present invention, there is provided a first reconstruction step of performing a reconstruction process including a successive approximation process on X-ray projection data acquired by imaging with a predetermined scan locus to generate a first image; A forward projection data generation step for generating forward projection data by forward projecting one image, and an analytical reconstruction process in which a desired reconstruction condition is set are performed on the forward projection data, and the final image is reconstructed. A scan trajectory in which less artifacts are mixed in the reconstruction process of the second reconstruction means in the forward projection data generation step than in the imaging in the forward projection data generation step, The reconstruction calculation method is characterized by forwardly projecting a first image.

  According to a third aspect of the present invention, there is provided a scanner that irradiates a subject with X-rays from a plurality of angular directions around the subject, collects X-ray projection data transmitted through the subject, and X-ray projection data collected by the scanner. An X-ray CT apparatus comprising: a reconstruction calculation device that reconstructs an X-ray CT image based on the X-ray projection data acquired by imaging with a predetermined scan locus. A first reconstruction unit that performs a reconstruction process including a successive approximation process to generate a first image; a forward projection data generation unit that generates forward projection data by forward projecting the first image; A second reconstruction unit configured to perform an analytical reconstruction process in which a desired reconstruction condition is set on the forward projection data and reconstruct a final image, and the forward projection data generation unit includes: Compared to the shooting, the second re- Scan locus contaminated artifacts in the reconstructed process is less adult means is an X-ray CT apparatus characterized by forward projecting the first image.

  ADVANTAGE OF THE INVENTION According to this invention, the reconstruction calculation apparatus, the reconstruction calculation method, and X-ray CT apparatus which can produce the X-ray CT image effective for medical diagnosis more rapidly and correctly can be provided.

External view showing the entire configuration of the X-ray CT apparatus 1 Hardware block diagram of X-ray CT apparatus 1 Explanatory drawing about single row detector 205A and multi-row detector 205B Illustration of normal scan and spiral scan Flowchart explaining the basic processing flow of the present invention The figure explaining the 1st reconstruction process, forward projection, and 2nd reconstruction process by the reconstruction calculating apparatus 45 Flowchart for explaining the first embodiment of the present invention. Flowchart for explaining the second embodiment of the present invention. Flowchart for explaining the third embodiment of the present invention.

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

First, with reference to FIGS. 1-2, the structure of the X-ray CT apparatus 1 of this Embodiment is demonstrated.
In the present embodiment, the X-ray CT apparatus 1 is a rotation-rotation method (Rotate-Rotate method) in which an X-ray source and an X-ray detector rotate together while irradiating a wide fan beam that covers the entire subject. ), An electron beam scanning method (Scanning Electron Beam method) in which the electron beam is applied to the target electrode while being electrically deflected, and other methods, but the present invention can be applied to any type of X-ray CT apparatus. It is.

  As shown in FIGS. 1 and 2, the X-ray CT apparatus 1 includes a scanner 2, a bed 3, and an operation unit 4. The X-ray CT apparatus 1 obtains projection data from each angle (view) direction around the subject by carrying the subject 6 fixed on the bed 3 into the opening of the scanner 2 and scanning.

  As shown in FIG. 2, the scanner 2 includes an X-ray generator (X-ray source) 201, a high-voltage generator 22, an X-ray controller 202, a collimator 203, a collimator controller 204, an X-ray detector 205, a preamplifier 206, The AD converter 207, the rotating plate 24, the drive device 208, the drive transmission system 210, the scanner control device 23, and the central control device 21 are configured.

  The X-ray generator 201 is an X-ray source, and continuously or intermittently irradiates the subject 6 with X-rays having an intensity according to the tube voltage / tube current applied / supplied from the high voltage generator 22. . The X-ray controller 202 controls the high voltage generator 22 according to the tube voltage and tube current determined by the central controller 21.

  The collimator 203 is configured to irradiate the subject 6 with X-rays radiated from the X-ray generator 201 in the shape of, for example, a cone beam (conical or pyramidal beam), and is controlled by the collimator controller 204. The X-rays transmitted through the subject 6 enter the X-ray detector 205.

  The X-ray detector 205 includes, for example, about 1000 X-ray detection element groups configured by, for example, a combination of a scintillator and a photodiode in the channel direction (circumferential direction), for example, about 1-320 in the column direction (body axis direction). They are arranged and arranged so as to face the X-ray generator 201 with the subject 6 interposed therebetween. The X-ray detector 205 detects X-rays radiated from the X-ray generator 201 and transmitted through the subject 6, amplifies the detected X-ray projection data by the preamplifier 206, and outputs the amplified X-ray projection data to the A / D converter 207. .

  The A / D converter 207 collects X-ray projection data detected by each X-ray detection element of the X-ray detector 205 and amplified by the preamplifier 206, converts it into a digital signal, and converts it into a digital signal so that the arithmetic unit 44 of the operation unit 4 is used. Output to.

  On the rotating plate 24, an X-ray generator 201, a collimator 203, an X-ray detector 205, a preamplifier 206, and the like are mounted. The rotating plate 24 is rotated by the driving force transmitted through the drive transmission system 210 from the driving device 208 controlled by the scanner control device 23.

  The central controller 21 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. The central control device 21 controls the X-ray tube control device 202, the collimator control device 204, the scanner control device 23, the arithmetic device 44 of the operation unit 4, and controls the bed control device 301 in the bed 3. .

  The bed 3 shown in FIG. 1 and FIG. 2 is moved to a predetermined height position, body axis position, and body width position under the control of the bed control device 301. Thereby, the subject 6 is carried into and out of the X-ray irradiation space of the scanner 2.

  The operation unit 4 includes a display device 41, an input device 42, a storage device 43, and an arithmetic device 44. The arithmetic device 44 includes a reconstruction arithmetic device 45 and an image processing device 46. The operation unit 4 is connected to the scanner 2. Each unit of the display device 41, the input device 42, the storage device 43, and the arithmetic device 44 is controlled by the central control device 21 in the scanner 2.

  The display device 41 includes a display device such as a liquid crystal panel and a CRT monitor, and a logic circuit for executing display processing in cooperation with the display device, and is connected to the arithmetic device 44. The display device 41 displays a reconstructed image (X-ray CT image) output from the arithmetic device 44 under control of the central controller 21 and the arithmetic device 44 and various information handled by the central controller 21.

  The input device 42 includes, for example, an input device such as a keyboard, a mouse, and a numeric keypad, and various switch buttons. The input device 42 outputs various instructions and information input by the operator to the arithmetic device 44. The computing device 44 outputs various instructions and information input from the input device 42 to the central control device 21. The operator operates the X-ray CT apparatus 1 interactively using the display device 41 and the input device 42.

  The storage device 43 is configured by a hard disk or the like, and is connected to the arithmetic device 44. The storage device 43 stores an image output from the arithmetic device 44, a program necessary for processing to be described later, and various data.

The arithmetic device 44 is constituted by a CPU, a ROM, a RAM, and the like, and controls the entire operation unit 4 according to the control of the central control device 21.
The reconstruction calculation device 45 acquires X-ray projection data output from the A / D converter 207 in the scanner 2 under the control of the central control device 21, and performs image reconstruction processing. That is, the reconstruction calculation device 45 performs preprocessing such as log conversion and calibration on the acquired X-ray projection data, and further performs processing (see FIG. 5) described later on the X-ray projection data. Thus, an X-ray CT image of the subject 6 is reconstructed. The reconstructed X-ray CT image is stored in the storage device 43 and displayed on the display device 41. The image processing device 46 is a device that performs an image processing process desired by the operator on the X-ray CT image generated by the reconstruction calculation device 45.

Next, the X-ray detector 205 and the X-ray beam will be described with reference to FIG.
The X-ray detector 205 includes a single-row detector 205A as shown in FIG. 3A and a multi-row detector 205B as shown in FIG.
The single-row detector 205A is a detector in which detection elements are arranged in one dimension (one row) in the rotating direction of the rotating plate 24, and is also called a one-dimensional detector or a single slice.
On the other hand, the multi-row detector 205B is a detector in which detection elements are arranged two-dimensionally in the rotation direction and the rotation axis direction of the rotating plate 24, and is also called a two-dimensional detector or multislice.

  In a single slice CT apparatus equipped with the single-row detector 205A, an X-ray beam spreading in a fan shape is emitted from the X-ray generator 201. In the multi-slice CT apparatus equipped with the multi-row detector 205B, an X-ray beam spreading in a cone shape or a pyramid shape having a cone angle corresponding to the number of rows of the multi-row detector 205B is irradiated from the X-ray generation device 201. Is done.

Next, the scanning method will be described with reference to FIG.
In the X-ray CT apparatus 1, X-ray irradiation is performed around the subject 6 placed on the bed 3 while the rotating plate 24 rotates. At this time, as shown in FIG. 4A, during the rotation of the rotating plate 24, the position of the bed 3 is not changed, and the X-ray source 201 rotates around the subject 6 so as to draw a circular orbit. This is called normal scan or axial scan. On the other hand, as shown in FIG. 4B, the position of the bed 3 is moved in the body axis (z) direction during the rotation of the rotating plate 24 so that the X-ray source 201 draws a spiral trajectory around the subject 6. Shooting that goes around is called helical scanning or helical scanning.

  The X-ray detector 205 of the X-ray CT apparatus 1 of the present invention is a multi-slice CT apparatus equipped with a multi-row detector 205B shown in FIG. When imaging the subject 6, the scanner 2 uses the multi-row detector 205B to acquire X-ray projection data by the normal scan of FIG. 4A or the spiral scan of FIG. 4B. The reconstruction calculation device 45 generates a first image by repeating forward projection (reprojection) and backprojection on a CT image analytically reconstructed by the FBP method or the like, as will be described later. The forward projection process is performed again after one image is generated, and the second reconstruction process is performed under desired reconstruction conditions using the forward projection data obtained by the forward projection process. In the forward projection process for generating the first image, the reconstruction calculation device 45 performs scanning using the single-row detector 205A or the multi-row detector 205B having a smaller number of rows than that at the time of imaging.

  Hereinafter, the operation of the X-ray CT apparatus 1 will be described with reference to the flowchart of FIG. 5 and FIG. 6.

  The central control device 21 of the X-ray CT apparatus 1 according to the present embodiment performs processing according to the procedure of the flowchart shown in FIG. 5 when performing imaging and image reconstruction of the subject 6. That is, the central control device 21 reads a program and data related to the processing from the storage device 43, and executes processing according to the following procedure based on the program and data.

  First, the central control device 21 accepts setting of photographing conditions and reconstruction conditions. Examples of imaging conditions include bed movement speed, tube current, tube voltage, collimator conditions, helical pitch, and slice position. As reconstruction conditions, reconstruction FOV, reconstruction center position, reconstruction pitch (slice thickness), region of interest, reconstruction image matrix size, reconstruction filter function, maximum number of iterations of successive approximation processing, convergence condition, and the like. (Step S1).

  When an imaging condition or the like is input from the input device 42 of the operation unit 4, the central processing unit 21 then sends a control signal necessary for imaging to the X-ray control device based on the imaging condition or the like input in step S1. 202, to the bed control device 301 and the scanner control device 23. Further, the central processing unit 21 issues a shooting start signal and starts shooting (step S2).

  When imaging is started, a control signal is sent to the high voltage generator 22 by the X-ray controller 202 and a high voltage is applied to the X-ray generator 201. The X-ray generator 201 irradiates the subject 6 with X-rays. At this time, a control signal is sent from the scanner control device 23 to the drive device 208, and the X-ray generator 201, the X-ray detector 205, the preamplifier 206, and the like circulate around the subject 6. On the other hand, the bed 3 on which the subject 6 is placed by the bed control device 301 is stationary at a predetermined slice position during normal scanning, and is translated in the body axis direction at a predetermined helical pitch during helical scanning.

  X-rays emitted from the X-ray generator 201 are limited in irradiation area by the collimator 203, absorbed (attenuated) by each tissue in the subject 6, pass through the subject 6, and detected by the X-ray detector 205. Is done. X-rays detected by the X-ray detector 205 are converted into current, amplified by the preamplifier 206, converted into digital data by the A / D converter 207, and output to the arithmetic unit 44 of the operation unit 4. The arithmetic unit 44 performs log conversion and calibration on the acquired digital data to obtain projection data. This projection data is subject to reconstruction computation by the reconstruction computation device 45.

When acquiring the projection data, the reconstruction calculation device 45 first performs a first reconstruction process to obtain a first reconstruction image (hereinafter referred to as a first image) (step S3).
In the first reconstruction process, the reconstruction calculation device 45 performs a reconstruction process including a successive approximation method. That is, the reconstruction calculation device 45 gives a predetermined initial image to the projection data and starts the reconstruction process by the successive approximation method, or in advance for the projection data in order to speed up the calculation process. An initial image is reconstructed by an analytical reconstruction method, and after confirming noise and artifacts in the initial image, reconstruction processing by a successive approximation method is started.

  As an analytical reconstruction method for obtaining an initial image, any method such as a Radon transform method, a Fourier transform method, a filtered back projection method (FBP method) or the like may be used. In the case of a multi-slice CT apparatus having 205, for example, a filter-corrected 3D backprojection process based on the Weighted Feldkamp method or a process obtained by improving this is suitable. When filter-corrected 3D backprojection processing based on the Weighted Feldkamp method is adopted, reconstruction processing is performed in consideration of the cone angle, so the number of iterations until the successive approximation processing converges is reduced, and the calculation time is reduced. it can. A constant value image may be used as the initial image. However, when a constant value image is used, since the likelihood for projection data is often low, the convergence of the successive approximation process is delayed, and the computation time is long. For this reason, from the viewpoint of speeding up the processing in step S3, it is desirable to employ an initial image with a high likelihood for the projection data as much as possible.

  The flow up to this point will be described with reference to FIG. FIG. 6A shows a method for correcting image data in the successive approximation process, and FIG. 6B shows a method for correcting projection data in the successive approximation process.

  In the method of correcting the image data shown in FIG. 6A, the reconstruction calculation device 45 reconstructs an image based on the projection data 61 obtained in steps S1 to S2. For example, a reconstructed image 62 is calculated for the projection data 61 by the FBP method, and the successive approximation process 63 is performed using this as an initial image. In the successive approximation process 63, the FBP reconstructed image 62 is reprojected to obtain reprojection data 64, the correction component projection data 65 is estimated from the projection data 61 and the reprojection data 64, and the estimated correction component projection is performed. Data 65 is reconstructed to generate a corrected component image 66, and the reconstructed image 62 used in the reprojection process is corrected to obtain a corrected image (first image 67). The above-described reprojection processing → reconstruction processing → image correction processing is repeatedly executed, and when the convergence condition is satisfied, the successive approximation processing 63 is terminated to obtain a corrected image (first image 67).

  Further, in the method of updating the projection data shown in FIG. 6B, the reconstruction calculation device 45 uses the projection data 71 obtained in step S1 to step S2 (the same as the projection data 61 in FIG. 6A). Based on this, after calculating the FBP reconstructed image 72 (the same as the FBP reconstructed image 62 in FIG. 6A), a successive approximation process 73 is further performed. In the successive approximation process 73, the FBP reconstructed image 72 is reprojected to obtain the reprojection data 74, the correction component projection data 75 is estimated from the projection data 71 and the reprojection data 74, and obtained by the reprojection process. The reprojection data 74 is corrected by the correction component projection data 75 to generate corrected projection data 76, which is reconstructed to obtain a corrected image (first image 77). The above reprojection process → projection data correction process → reconstruction process is repeatedly executed, and when the convergence condition is satisfied, the successive approximation process 73 is terminated to obtain a corrected image (first image 77).

  As the successive approximation processing 63 and 73, ML (maximum likelihood estimation: Maximum Likelihood) method, MAP (maximum posterior probability: Maximum a Osteriar) method, WLS (weighted least squares) method, PWLS (penalized weighting) A known method such as the least square method: Penalized Weighted Least Squares (SIRT) method or SIRT (Simultaneous Reconstruction Technique) method may be used. Further, a speed-up method such as OS (Ordered Subset) or SPS (Separable Parabolic Surrogate) may be applied to these successive approximation processes.

  As described above, when the first reconstruction process (step S3) is completed, the first images 67 and 77 from which cone beam artifacts, noise, and the like are removed are obtained. For example, in step S2 (imaging), a 64-row multi-slice CT apparatus is used to obtain projection data 61 and 71 by performing high-speed spiral scanning with a helical pitch of about 80 or normal scanning with a slice thickness of about 0.625 mm. And In this case, cone beam artifacts occur in the FBP reconstructed images 62 and 72 described above. Such cone beam artifacts are removed by the successive approximation processes 63 and 73 described above.

  In the first images 67 and 77 obtained by the successive approximation processes 63 and 73, cone beam artifacts and the like are removed and the image quality is improved. However, in the successive approximation process, correction data is estimated based on actual projection data. The influence of preset reconstruction conditions such as a reconstruction filter, a reconstruction FOV, and a reconstruction center cannot be obtained.

Therefore, the reconstruction calculation device 45 executes the second reconstruction process, and generates a final image by an analytical reconstruction process to which a desired reconstruction condition is applied while maintaining the image quality of the first images 67 and 77. .
This second reconstruction process is performed on the forward projection data 68 and 78 of the first images 67 and 77 with improved image quality.

  That is, the reconstruction calculation device 45 performs forward projection processing on the first images 67 and 77 obtained in step S3. In this forward projection process, a virtual scan is performed with a scan trajectory with less artifacts compared to the scan trajectory in step S2 (step S4).

  Here, a scan trajectory with less artifacts will be described. Artifacts appear in an image obtained by reconstructing the forward projection data 68 and 78 (final images 69 and 79 obtained after the second reconstruction process). Therefore, the forward projection data 68 and 78 generated here may be generated so that the final images 69 and 79 do not include artifacts as much as possible.

  In order to prevent the final images 69 and 79 from including artifacts, forward projection processing using a detector 205 having a narrow cone angle and a scanning method may be performed. Further, it is preferable to combine reconstruction processing suitable for the obtained forward projection data 68 and 78.

  That is, the virtual detector used in the forward projection process is a single-row detector 205A (single slice), or has a smaller number of columns than the X-ray detector (for example, 64 rows) used at the time of imaging. It is desirable to use a column detector 205B (for example, about 4 to 16 columns). In addition, since the image quality is improved when the number of data related to the image to be reconstructed is large, the scanning method is normal scanning for the single-row detector 205A (single slice), and low speed for the multi-row detector 205B. A spiral scan is desirable. In particular, it is known that the filter-corrected 2D reconstruction method performed by a conventional single slice CT apparatus can obtain a mathematically exact solution (accurate solution).

  Therefore, specifically, as in Example (1) to be described later, the forward projection process is a normal scan (two-dimensional forward projection) using a single-row detector 205A (single slice), and the order obtained thereby. If the filter-corrected 2D reconstruction method is applied to the projection data 68 and 78, an accurate solution (final images 69 and 79) can be obtained. The calculation speed is also high.

  Even with an approximate solution, the cone angle can be reduced by using a multi-row detector 205B having a smaller number of rows as in the embodiments (2) and (3) described later, and a slower spiral scan can be achieved. If it is (that is, a dense scan in the body axis direction), more forward projection data used for reconstruction calculation can be obtained.

As a specific example of the above-described forward projection process and the second reconstruction process, an example (1) is first shown in FIG.
In the embodiment (1) of FIG. 7, the reconstruction calculation device 45 performs setting of shooting conditions (step S11), shooting according to shooting conditions (step S12), and shooting in the same manner as in steps S1 to S3 described above. Filter correction 3D backprojection processing based on the obtained projection data (step S13) and successive approximation processing based on the initial image obtained in step S13 (step S14) are performed.

  Next, the reconstruction calculation device 45 performs 2D forward projection processing on the first images 67 and 77 obtained in step S14. The 2D forward projection processing is forward projection by single slice (single row detector) and normal scanning (step S15).

  In this case, the filter correction 2D back projection process can be used in the second reconstruction process (step S16). By adopting the filter correction 2D backprojection process, an accurate solution can be obtained in which there is no contradiction between the forward projection data 68 and 78 and the reconstructed image (final images 69 and 79). Also, if a filter having a desired frequency characteristic is applied as a reconstruction filter for the filter correction 2D backprojection process, or reconstruction is performed with a desired FOV, the first image 67, generated in step S14, The final images 69 and 79 effective for diagnosis, which are different from 77, can be generated.

For example, when the imaging region is the head, since it is desired to identify a small lesion and a small lesion, an image with high resolution may be generated using a reconstruction filter that emphasizes higher frequency components. Further, when the imaging region is the abdomen, the subject size is large, so a filter that suppresses high frequency components may be used for the purpose of reducing noise.
Therefore, according to the embodiment (1), an accurate solution for the forward projection data 68 and 78 of the first images 67 and 77 whose image quality is improved by the successive approximation process and a characteristic effective for diagnosis (final) Images 69 and 79) can be obtained.

FIG. 8 shows an embodiment (2) of the forward projection process and the second reconstruction process described above.
In the embodiment (2) of FIG. 8, the reconstruction calculation device 45 performs the shooting condition setting (step S21), shooting according to the shooting conditions (step S22), and shooting in the same manner as in steps S1 to S3 described above. A filter-corrected 3D backprojection process (step S23) based on the obtained projection data and a successive approximation process (step S24) based on the initial image obtained in step S23 are performed.

  Next, the reconstruction calculation device 45 performs 3D forward projection processing on the first images 67 and 77 obtained in step S24. The 3D forward projection process is forward projection by multi-row slice (multi-row detector) and spiral scan (step S25). In this embodiment (2), for example, the number of columns is about 4 and the helical pitch is about 5.

In this case, the spiral correction filter correction 2D reconstruction process can be used in the second reconstruction process (step S26).
Compared to the case of the embodiment (1), since there are a plurality of detector columns in the forward projection process, a cone angle is generated and an approximate solution is obtained. The final images 69 and 79 to which desired reconstruction conditions are applied can be acquired. Further, since the reconstruction process is two-dimensional, the calculation time can be performed as fast as in the embodiment (1). In particular, when the reconstruction pitch (slice thickness; interval in the body axis direction of the image plane) is narrowed, the calculation can be performed at a higher speed than in the embodiment (1). Further, similarly to the embodiment (1), since a desired reconstruction condition (frequency characteristics, etc.) can be set during the 2D reconstruction process in step S26, it is different from the first images 67 and 77 generated in step S24. The final images 69 and 79 effective for diagnosis can be generated.

FIG. 9 shows an embodiment (3) of the above-described forward projection process and the second reconstruction process.
In the embodiment (3) of FIG. 9, the reconstruction calculation device 45 performs the shooting condition setting (step S31), shooting according to the shooting conditions (step S32), and shooting, as in steps S1 to S3 described above. Filter correction 3D backprojection processing based on the obtained projection data (step S33) and successive approximation processing based on the initial image obtained in step S33 (step S34) are performed.

  Next, the reconstruction calculation device 45 performs 3D forward projection processing on the first images 67 and 77 obtained in step S34. In this embodiment (3), for example, the number of columns is about 16 and the helical pitch is about 20.

In this case, a weighted filter correction 3D reconstruction process based on the weighted Feldkamp method can be used in a second reconstruction process described later (step S36).
By making the second reconstruction processing three-dimensional, the calculation time is slower than in the embodiments (1) and (2), but the number of data to be referred to generate one image is increased. The image quality is better than the final images 69 and 79 obtained by the example (2). Similarly to the embodiments (1) and (2), the first image 67, 77 generated in step S34 can be reconstructed with desired shooting conditions (frequency characteristics, etc.) during the 3D reconstruction process in step S36. It is possible to generate a diagnostically effective image different from the above.

  Note that the reconstruction process in the procedure of the present embodiment does not have to be applied to all captured images, and is applied to an image including a part to be examined in more detail or an image to be filmed (exposure recorded). You may make it apply only to. In this case, it is possible to efficiently generate an image for a doctor or the like who makes a diagnosis, such as giving priority to image quality improvement for an image including a part to be examined in detail, and giving priority to calculation processing speed for other images.

  As described above, the X-ray CT apparatus 1 of the present invention uses the multi-row detector and the projection data acquired by imaging using a predetermined scan locus, and includes the first reconstruction process including the successive approximation process. Thus, the first images 67 and 77 are generated. Thereafter, forward projection is performed on the generated first images 67 and 77 with a scan trajectory with less artifacts compared to photographing. Then, based on the forward projection data 68 and 78, a photographing condition such as a desired frequency condition is set, for example, and a second reconstruction process using an analytical method is performed to generate final images 69 and 79.

  Therefore, by applying successive approximation processing (first reconstruction processing) to projection data obtained by imaging with multi-row and high-speed scanning, it is possible to reduce noise and cone beam artifacts while performing imaging with less burden on the patient. A simple image. In order to reconstruct an image effective for diagnosis based on such a high-quality image, the forward projection process was performed with a scan trajectory with less artifact mixing than that at the time of shooting, and image quality deterioration was suppressed and obtained. A desired reconstruction condition is applied to the forward projection data to perform a second reconstruction process by an analytical method. Therefore, an image effective for diagnosis expressing desired characteristics (frequency characteristics, reconstructed FOV, etc.) can be generated with good image quality.

  In particular, when a two-dimensional forward projection process (normal scan using a single-row detector) is adopted for the scan during the forward projection process, the filter-corrected 2D backprojection method, which is a complete algorithm, is used to accurately and accurately It becomes possible to obtain a final image having desired characteristics at high speed.

  In addition, when a multi-row detector having a smaller number of columns than that at the time of photographing is used for the scan at the time of forward projection processing and a spiral scan (three-dimensional forward projection processing) that is slower than at the time of photographing is employed, It is possible to perform reconstruction processing using filter correction reconstruction processing suitable for the pitch. For example, if the filter correction 2D reconstruction process is performed, it is possible to perform calculations at high speed, and if the filter correction 3D reconstruction process is performed, the image quality can be improved.

In the above-described embodiment, the scan trajectory of the forward projection process is a normal scan or a spiral scan. However, the scan trajectory is not limited to this, and a combination of a plurality of scans such as a combination of a normal scan and a line scan. Also good. In any scan or combination thereof, forward projection is performed such that the cone angle is small and the number of data related to the reconstructed image is large.
The second reconstruction process is not limited to the filter-corrected back projection method, and a filter superposition type reconstruction method such as a Fourier reconstruction method may be used.

  The preferred embodiments of the reconstruction calculation device, the reconstruction calculation method, and the X-ray CT apparatus according to the present invention have been described above, but the present invention is not limited to the above-described embodiments. For example, in the above-described embodiment, the gantry type X-ray CT apparatus has been described, but a C-arm type X-ray CT apparatus may be used. In addition, it is obvious for those skilled in the art that various changes or modifications can be conceived within the scope of the technical idea disclosed in the present application, and these naturally belong to the technical scope of the present invention. It is understood.

DESCRIPTION OF SYMBOLS 1 ... X-ray CT apparatus 2 ... Scanner 21 ... Central controller 201 ... X-ray generator (X-ray source)
205 ... X-ray detector 3 ... bed 4 ... operating unit 41 ... display device 42 ... operating device 43 ... storage device 44 ... calculation Device 45... Reconstruction arithmetic device 46... Image processing device 6... Subject 205 A... Single row detector 205 B.

Claims (7)

First reconstruction means for performing reconstruction processing including successive approximation processing on X-ray projection data acquired by imaging with a predetermined scan locus to generate a first image;
Forward projection data generation means for generating forward projection data by forward projecting the first image;
A second reconstruction unit configured to perform an analytical reconstruction process in which a desired reconstruction condition is set on the forward projection data and reconstruct a final image;
The forward projection data generation means includes:
A reconstruction calculation device that projects the first image in a forward direction with a scan trajectory in which artifacts in the reconstruction processing of the second reconstruction unit are less mixed than in the imaging. The forward projection data generation means performs a two-dimensional forward projection process on the first image by a single row detector and a normal scan,
2. The second reconstructing means applies a filter-corrected two-dimensional backprojection process under a desired reconstruction condition to the forward projection data acquired by the two-dimensional forward projection process. The reconstruction calculation device described in 1. The forward projection data generation means performs a three-dimensional forward projection process on the first image by a multi-row detector having a smaller number of rows than that at the time of shooting and a low-speed spiral scan,
2. The second reconstruction unit applies a filter-corrected two-dimensional backprojection process under a desired reconstruction condition to the forward projection data acquired by the three-dimensional forward projection process. The reconstruction calculation device described in 1. The forward projection data generating means performs a three-dimensional forward projection process on the first image by a multi-row detector having a smaller number of rows than that at the time of photographing and a low-speed spiral scan,
2. The second reconstructing means applies a filter-corrected three-dimensional backprojection process under a desired reconstruction condition to the forward projection data acquired by the three-dimensional forward projection process. The reconstruction calculation device described in 1. The reconstruction operation device according to any one of claims 1 to 4, wherein the reconstruction condition includes a frequency characteristic condition. A first reconstruction step of performing a reconstruction process including a successive approximation process on the X-ray projection data acquired by imaging with a predetermined scan locus to generate a first image;
A forward projection data generation step of generating forward projection data by forward projecting the first image;
A second reconstruction step of performing an analytical reconstruction process in which a desired reconstruction condition is set on the forward projection data, and reconstructing a final image,
In the forward projection data generation step,
A reconstruction calculation method characterized by forwardly projecting the first image with a scan trajectory with less artifacts in the reconstruction processing of the second reconstruction means compared to the imaging. A X-ray CT image based on X-ray projection data collected by a scanner that irradiates a subject with X-rays from a plurality of angular directions around the subject and collects X-ray projection data transmitted through the subject An X-ray CT apparatus comprising a reconstruction calculation device for reconfiguring
The reconstruction arithmetic device is
A first reconstruction means for performing a reconstruction process including a successive approximation process on the X-ray projection data acquired by imaging with a predetermined scan locus, and generating a first image;
Forward projection data generation means for generating forward projection data by forward projecting the first image;
A second reconstruction unit configured to perform an analytical reconstruction process in which a desired reconstruction condition is set on the forward projection data and reconstruct a final image;
The forward projection data generation means includes:
An X-ray CT apparatus for forwardly projecting the first image with a scan trajectory with less artifacts in the reconstruction processing of the second reconstruction means compared to the imaging.

Images