JP2016059509A - Image processor, x-ray ct apparatus and image processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processor for improving the calculation accuracy of a linear attenuation coefficient of a restored image being a cross-sectional image of a subject, and to provide an X-ray CT apparatus and an image processing method.SOLUTION: According to an embodiment, the image processor is an image processor for executing image processing of an image based on the energy of a radiation ray transmitting a subject, and includes a projection data processing part, a first reconfiguration part, a second reconfiguration part, and a generation part. The projection data processing part generates second projection data with a value obtained by performing a prescribed calculation as a pixel value on the basis of a pixel value of a prescribed range in a prescribed direction of first projection data based on the photon of the energy of the radiation ray. The first reconfiguration part reconfigures the first projection data to generate a first reconfiguration image. The second reconfiguration part reconfigures the second projection data to generate a second reconfiguration image. The generation part generates a restored image by converting the first reconfiguration image on the basis of the second reconfiguration image.SELECTED DRAWING: Figure 2

Description

  Embodiments described herein relate generally to an image processing apparatus, an X-ray CT apparatus, and an image processing method.

  In recent years, development of silicon-based photomultipliers has become active, and radiation detection apparatuses such as X-ray CT (Computed Tomography) apparatuses using photomultipliers have been developed. As an example of such an X-ray CT apparatus, there is a spectral CT apparatus that detects a spectrum indicated by the number of photons for each X-ray energy transmitted through a subject.

  In such a spectral CT apparatus, for the line attenuation coefficient that is a pixel value of a cross-sectional image obtained by reconstructing from a sinogram of a specific energy band of interest, it is suppressed that the calculation accuracy of the line attenuation coefficient is reduced due to noise. In order to do this, there is a technique for performing the following treatment. That is, by using a cross-sectional image obtained by reconstructing from a sinogram of an energy band wider than the specific energy band described above, there is a technique for obtaining a cross-sectional image composed of a linear attenuation coefficient in which the influence of noise is reduced. Proposed.

  However, in such a spectral CT apparatus, when imaging is performed under a condition where the dose of irradiated X-rays is low, a subject having a high linear attenuation coefficient or a subject having a large thickness, or a specific energy band of interest May be reduced, a channel may be generated in which the number of photons detected by the detector is zero. In such a case, there is a problem that the calculation accuracy of the linear attenuation coefficient constituting the cross-sectional image of the subject is lowered.

US Patent Application Publication US2013 / 0108013

  The present invention has been made in view of the above, and provides an image processing apparatus, an X-ray CT apparatus, and an image processing method that improve the calculation accuracy of a linear attenuation coefficient of a restored image that is a cross-sectional image of a subject. With the goal.

  An image processing apparatus according to an embodiment is an image processing apparatus that performs image processing on an image based on energy of radiation that has passed through a subject, and includes a projection data processing unit, a first reconstruction unit, and a second reconstruction unit. A configuration unit and a generation unit are provided. The projection data processing unit is a second projection data in which a pixel value is a value obtained by performing a predetermined calculation based on a pixel value in a predetermined range in a predetermined direction of the first projection data based on photons of radiation energy transmitted through the subject. Is generated. The first reconstruction unit reconstructs the first projection data and generates a first reconstructed image. The second reconstruction unit reconstructs the second projection data and generates a second reconstructed image. The generation unit generates a restored image by converting the first reconstructed image based on the second reconstructed image.

1 is an overall configuration diagram of an X-ray inspection apparatus according to a first embodiment. It is a figure which shows an example of the block configuration of the image processing part of 1st Embodiment. It is a figure explaining a sinogram. It is a figure which shows an example of the spectrum of the energy detected by the specific channel of the detector. It is a figure which shows the example of a subject sinogram. It is a figure which shows the example of the reconstruction image reconfigure | reconstructed with the photon of a specific energy band. It is a figure which shows the example of the reconstruction image reconfigure | reconstructed by 2nd projection data. It is a figure which shows the example of the decompression | restoration image after an image process. 3 is a flowchart illustrating an example of an operation of an image processing unit according to the first embodiment. It is a figure which shows the block configuration of the projection data processing part of the modification 1 of 1st Embodiment. It is a flowchart which shows an example of operation | movement of the image process part of the modification 1 of 1st Embodiment. It is a figure which shows the block configuration of the projection data processing part of the modification 2 of 1st Embodiment. It is a flowchart which shows an example of operation | movement of the image process part of the modification 2 of 1st Embodiment. It is a figure which shows an example of the block configuration of the image processing part of 2nd Embodiment. It is a figure which shows an example of the display content of the display apparatus of 2nd Embodiment. It is a flowchart which shows an example of operation | movement of the image process part of 2nd Embodiment.

  Hereinafter, an image processing apparatus, an X-ray CT apparatus, and an image processing method according to an embodiment of the present invention will be described in detail with reference to the drawings. Moreover, in the following drawings, the same code | symbol is attached | subjected to the same part. However, since the drawings are schematic, a specific configuration should be determined in consideration of the following description.

(First embodiment)
FIG. 1 is an overall configuration diagram of the X-ray inspection apparatus according to the first embodiment. An overview of the overall configuration of the X-ray inspection apparatus 1 will be described with reference to FIG.

  As shown in FIG. 1, the X-ray inspection apparatus 1, which is an example of an X-ray CT apparatus, transmits X-rays to the subject 40 and detects it as a spectrum indicated by the number of photons for each energy. A spectral CT apparatus, a photon counting CT apparatus, or the like that obtains a cross-sectional image of the projected cross section 41. As shown in FIG. 1, the X-ray inspection apparatus 1 includes a gantry device 10, a bed device 20, and a console device 30 (image processing device).

  The gantry device 10 is a device that irradiates and transmits the subject 40 with X-rays and detects the above-described spectrum. The gantry device 10 includes an X-ray tube 11, a rotating frame 12, a detector 13, an irradiation control unit 14, a gantry driving unit 15, and a data collection unit 16.

  The X-ray tube 11 is a vacuum tube that generates X-rays with a high voltage supplied from the irradiation control unit 14, and irradiates the subject 40 with the X-ray beam 11 a. The spectrum indicated by the number of photons for each X-ray energy irradiated from the X-ray tube 11 is determined by the tube voltage of the X-ray tube 11, the tube current, and the type of target (for example, tungsten) used for the radiation source. . When the X-rays irradiated from the X-ray tube 11 pass through the subject 40, the number of photons of each energy is decreased and the spectrum is changed according to the state of the substance constituting the subject 40.

  The rotating frame 12 is a ring-shaped support member that supports the X-ray tube 11 and the detector 13 so as to face each other with the subject 40 interposed therebetween.

  The detector 13 is a detector that detects, for each channel, the number of photons for each X-ray energy irradiated from the X-ray tube 11 and transmitted through the subject 40. That is, the detector 13 detects, for each channel, a spectrum indicated by the number of photons for each X-ray energy as shown in FIG. 4 described later. As shown in FIG. 1, the detector 13 detects a spectrum for each view while rotating in the circumferential direction of the rotating frame 12. Here, the view means an angle when a spectrum is detected by the detector 13 at every predetermined angle out of 360 degrees in the circumferential direction of the rotating frame 12. That is, when the detector 13 detects a spectrum every 2 °, 1 view = 2 °. The detector 13 includes a plurality of detection element arrays in which a plurality of detection elements are arranged in the channel direction (circumferential direction of the rotating frame 12) along the body axis direction of the subject 40 (Z-axis direction shown in FIG. 1). It is a two-dimensional array type detector arranged. Note that the detection element array of the detector 13 may be configured by a combination of a photocounting type detection element and an integration type detection element.

  The irradiation control unit 14 is a device that generates a high voltage and supplies the generated high voltage to the X-ray tube 11.

  The gantry driving unit 15 is a device that rotationally drives the rotary frame 12 to rotationally drive the X-ray tube 11 and the detector 13 on a circular orbit around the subject 40. The gantry driving unit 15 is not limited to a configuration that rotationally drives both the X-ray tube 11 and the detector 13. For example, the detector 13 may be configured such that detection elements are arranged for one turn in the circumferential direction of the rotating frame 12, and the gantry driving unit 15 may be configured to rotate only the X-ray tube 11. .

  The data collection unit 16 is a device that collects spectrum data indicated by the number of photons for each energy detected for each channel by the detector 13. Then, the data collection unit 16 performs amplification processing, A / D conversion processing, or the like on each of the collected spectrum data, generates a sinogram of a specific energy band desired (attention), and sends it to the console device 30. Output.

  The couch device 20 is a device on which the subject 40 is placed, and includes a couch driving device 21 and a top plate 22 as shown in FIG.

  The top plate 22 is a bed such as a bed on which the subject 40 is placed. The couch driving device 21 is a device that moves the subject 40 into the rotary frame 12 by moving the subject 40 placed on the top plate 22 in the body axis direction (Z-axis direction).

  The console device 30 is a device that accepts an operation of the X-ray inspection apparatus 1 by an operator and reconstructs a cross-sectional image from data collected by the gantry apparatus 10. As shown in FIG. 1, the console device 30 includes an input device 31, a display device 32, a scan control unit 33, an image processing unit 34, an image storage unit 35, and a system control unit 36. .

  The input device 31 is a device for an operator who operates the X-ray inspection apparatus 1 to input various instructions, and is a device that transmits various commands input by the operation to the system control unit 36. The input device 31 is, for example, a mouse, a keyboard, a button, a trackball, or a joystick.

  The display device 32 displays a GUI (Graphical User Interface) for accepting an operation instruction from the operator via the input device 31, or a reconstructed image (restored image, cross-sectional image) stored in the image storage unit 35 described later. It is a device that displays. The display device 32 is, for example, a CRT (Cathode Ray Tube) display, an LCD (Liquid Crystal Display), an organic EL (Organic Electro-Luminescence) display, or the like.

  The scan control unit 33 is a processing unit that controls operations of the irradiation control unit 14, the gantry driving unit 15, the data collection unit 16, and the bed driving device 21. Specifically, the scan control unit 33 performs X-ray scanning by continuously or intermittently irradiating X-rays from the X-ray tube 11 while rotating the rotating frame 12. For example, the scan control unit 33 performs helical scanning in which imaging is performed by continuously rotating the rotating frame 12 while moving the top plate 22, or imaging is performed by rotating the rotating frame 12 once around the subject 40, and then Then, the top plate 22 on which the subject 40 is placed is slightly shifted, and the rotating frame 12 is rotated once again to perform non-helical scanning for imaging.

  The image processing unit 34 is a processing unit that reconstructs a tomographic image of the subject from the sinogram received from the data collection unit 16. Details of the block configuration and operation of the image processing unit 34 will be described later.

  The image storage unit 35 is a functional unit that stores a cross-sectional image (restored image) generated by the reconstruction processing by the image processing unit 34. The image storage unit 35 is, for example, a storage device such as an HDD (Hard Disk Drive), an SSD (Solid State Drive), or an optical disk.

  The system control unit 36 is a processing unit that controls the entire X-ray inspection apparatus 1 by controlling operations of the gantry device 10, the couch device 20, and the console device 30. Specifically, the system control unit 36 controls the scan control unit 33 to control the collection operation of the spectrum data of the subject by the gantry device 10 and the couch device 20. The system control unit 36 also controls the tomographic image reconstruction process by controlling the image processing unit 34. Further, the system control unit 36 reads the cross-sectional image from the image storage unit 35 and causes the display device 32 to display the cross-sectional image.

  Although the data collection unit 16 generates a sinogram of a specific energy band desired (attention) from the collected spectrum data, the present invention is not limited to this. That is, the data collection unit 16 transmits the collected spectrum data to the image processing unit 34, and the image processing unit 34 generates a sinogram of a specific energy band desired (attention) from the spectrum data. Also good.

  FIG. 2 is a diagram illustrating an example of a block configuration of the image processing unit according to the first embodiment. FIG. 3 is a diagram for explaining a sinogram. FIG. 4 is an example of a spectrum of energy detected in a particular channel of the detector. FIG. 5 is a diagram illustrating an example of a subject sinogram. FIG. 6 is a diagram illustrating an example of a reconstructed image reconstructed with photons in a specific energy band. FIG. 7 is a diagram illustrating an example of a reconstructed image reconstructed with the second projection data. FIG. 8 is a diagram illustrating an example of a restored image after image processing. An outline of the configuration and operation of the image processing unit 34 according to the first embodiment will be described with reference to FIGS.

  As shown in FIG. 2, the image processing unit 34 includes a projection data processing unit 341, a first reconstruction unit 342, a second reconstruction unit 343, and a generation unit 344.

Projection data processing unit 341 from the data acquisition unit 16 receives the object sinogram is sinogram of the object 40 as the first projection data I _1, on the basis of the first projection data I _1, first in the form of a sinogram a processing unit that generates second projection data I _2. The first projection data I ₁ is a subject sinogram of a specific energy band desired (attention). Here, the sinogram refers to an image in which measured values for each view of the X-ray tube 11 and for each channel of the detector 13 are arranged as pixel values, like a sinogram 1001 shown in FIG. Among these, a sinogram generated from a spectrum (see FIG. 4) in which X-rays irradiated from the X-ray tube 11 are transmitted through the subject 40 and detected by the detector 13 is referred to as a subject sinogram. A sinogram generated from a spectrum detected by the detector 13 with X-rays passing through only air without placing the subject 40 is referred to as an air sinogram. The pixel values of the subject sinogram and the air sinogram are the number of photons detected as a measurement value by the detector 13. Further, since the detector 13 detects a spectrum indicated by the number of photons for each energy for each view and for each channel, the X-ray inspection apparatus 1 performs an X-ray scan for one round of the X-ray tube 11. A subject sinogram 1011 for each energy as shown in FIG. 5 can be obtained. In the example shown in FIG. 5, the spectrum is divided into four energy bands, and four object sinograms 1011a to 1011d are obtained for each energy band (hereinafter, also simply referred to as “for each energy”). In addition, although the example divided | segmented into four energy bands was shown in FIG. 5, it is not limited to this division | segmentation number.

As an example of a predetermined calculation for generating the second projection data I ₂ , the projection data processing unit 341 weights the pixel value of the first projection data I _{1 in} the channel direction as shown in the following formula (1), for example. second and projection data I ₂ to calculate the sum.

In equation (1), (x, y) indicates the data of the x th channel and the y th view, W _k is a real weight of 1 or more, and k is the shift amount in the channel direction. Yes, K and -K are the upper and lower limits of the range in the channel direction for calculating the weighted sum. W _k may be the same value regardless of the value of k, or may be larger as the value of k approaches 0.

As described above, when imaging is performed under a condition where the dose of X-rays irradiated from the X-ray tube 11 is low, a subject with a high linear attenuation coefficient or a subject with a large thickness is captured, or specific energy of interest When the band is narrow, a channel in which the number of photons detected by the detector 13 is 0 may be generated. In this case, the first projection data I ₁ includes a large number of pixel values (eg, “0”) indicating that the number of photons is zero. However, the projection data processing unit 341 calculates the weighted sum in the channel direction of the first projection data I ₁ by the above equation (1), so that the number of pixel values indicating that the number of photons is 0 is the first. Second projection data I ₂ that is smaller than the projection data I ₁ is generated. In any case, the first projection data I ₁ and the second projection data I ₂ are data based on the number of photons in the same energy band in the above-described spectrum (see FIG. 4).

Further, generation of the second projection data I ₂ are, from the viewpoint of reducing the number of pixel values that indicates that the number of photons in the second projection data I ₂ is 0, the addition with more data ( However, from the viewpoint of reducing the spatial resolution of a second reconstructed image (described later) reconstructed from the second projection data I ₂ as much as possible, addition (weighted sum) is performed using less data. It is preferable to do.

As described above, the calculation for obtaining the weighted sum in the channel direction of the first projection data I ₁ shown in Expression (1) is shown as an example of the predetermined calculation for generating the second projection data I ₂ . The present invention is not limited to this, and the second projection data I ₂ is generated using a non-linear filter such as a median filter or an epsilon filter as a predetermined calculation including addition of the pixel values of the first projection data I _1. Also good. By using such a nonlinear filter, it is possible to suppress a decrease in the spatial resolution of the second reconstructed image described later.

Further, the second projection data I ₂ may have a smaller number of pixels in the channel direction than the first projection data I ₁ . For example, the weighted sum of the number of pixels in the channel direction of the first projection data I ₁ is calculated every 10 channels, and the number of pixels in the channel direction of the second projection data I ₂ is calculated in the channel direction of the first projection data I ₁ . or as generating second projection data I ₂ so as to be 1/10 of the number of pixels.

Further, after calculating the weighted sum of the first projection data I _{1 according to} the above-described formula (1), the data may be subsampled at a predetermined channel interval to generate the second projection data I ₂ .

Further, in the above formula (1), the weighted sum of the first projection data I ₁ in the channel direction as the predetermined direction is calculated to be the second projection data I _2. However, the present invention is not limited to this. is not. For example, a weighted sum of the view direction, slice direction, time direction, or energy direction as a predetermined direction of the first projection data I ₁ may be calculated and used as the second projection data I ₂ . Further, the calculation processing of the weighted sum in each direction may be used in combination.

Here, the weighted sum of the first slice direction of the projection data I _1, a first weighted sum of projection data I ₁ generated in a plurality of projected section 41 of the Z-axis direction of the subject 40 of Figure 1. In this case, the pixel value of the first projection data I ₁ to be subjected to the weighted sum is the pixel value at the same position (x, y) in the plurality of first projection data I ₁ obtained in the slice direction. For example, it is possible to cross-sectional images of the object 40 when it is expected that the same image in the slice direction (Z-axis direction in FIG. 1), applies the weighted sum of the first slice direction of the projection data I _1. In addition, the weighted sum of the first time direction of the projection data I _1, is a weighted sum of the first projection data I ₁ obtained plurality in a time series in the same projected section 41 of the sample 40 of Figure 1. In this case, the pixel value of the first projection data I ₁ to be subjected to the weighted sum is the pixel value at the same position (x, y) in the plurality of first projection data I ₁ obtained in the time direction. For example, when the subject 40 is a human body and the organ corresponding to the projection section 41 moves, the weighted sum in the time direction of the first projection data I ₁ is not appropriate. Further, the weighted sum in the energy direction of the first projection data I ₁ is an object sinogram obtained from a spectrum of an energy band wider than this specific energy band including the above-mentioned desired (attention) specific energy band. It is a weighted sum for certain projection data.

The projection data processing unit 341, it is assumed to receive the first projection data I ₁ from the data acquisition unit 16, but is not limited thereto. For example, the data collection unit 16 stores the first projection data I ₁ that is the generated subject sinogram in the image storage unit 35, and the projection data processing unit 341 receives the first projection data I from the image storage unit 35. ₁ may be read and acquired.

Further, although the pixel value of the first projection data I ₁ that is the subject sinogram is the number of photons detected as the measurement value by the detector 13, it is not limited to this. For example, data acquired by a detector that detects the total energy of photons that have reached the detector 13 may be used as the pixel value of the first projection data I ₁ .

The first reconstruction unit 342 is a processing unit that reconstructs the first projection data I ₁ received from the data collection unit 16 and generates a first reconstructed image S ₁ . As a reconstruction method, for example, a filtered back projection method (FBP (Filtered Back Projection) method) which is an example of a back projection method can be used. Here, as a method for reconstructing the first projection data I _1, it will be described the case of adopting the FBP method. In this case, the first reconstruction unit 342 does not place the subject 40 and the reference projection data I ₀ that is an air sinogram generated from the spectrum detected by the detector 13 through which X-rays pass only air. It shall have.

Specifically, the first reconstruction unit 342 first calculates the integral value M ₁ of the linear attenuation coefficient from the first projection data I ₁ and the reference projection data I _{0 according} to the following equation (2).

In the case of I ₁ (x, y) = 0, the value of the integral value M ₁ (x, y) is calculated by the calculation using the upper expression of the expression (2) under the same condition as when I ₁ (x, y) ≠ 0. Becomes infinite and computation becomes impossible. Therefore, when I ₁ (x, y) = 0, the first reconstructing unit 342 replaces the first projection data I ₁ on the right side of the upper equation of Equation (2) with “1”. The integral value M ₁ (x, y) is calculated by This integral value M ₁ (x, y) is the subject along the path through which the X-rays emitted from the X-ray tube 11 reach the x-th channel of the detector 13 in the y-th view. It is a value obtained by integrating 40 linear attenuation coefficients.

Next, the first reconstruction unit 342 performs a one-dimensional Fourier transform in the channel direction on the calculated integral value M ₁ (x, y). Next, the first reconstruction unit 342 performs a filtering process in the frequency direction with a high-pass filter such as a ramp filter or a Shepp-Logan filter on the value subjected to the one-dimensional Fourier transform, thereby performing a one-dimensional inverse Fourier transform. do. The first reconstruction unit 342, the data in which the one-dimensional inverse Fourier transform, by adding backprojected into each all views, the first reconstructed image S ₁ is reconstructed sectional images Generate.

The second reconstruction unit 343 is a processing unit that reconstructs the second projection data I ₂ received from the projection data processing unit 341 and generates a second reconstructed image S ₂ . Here, as a method for reconstructing the second projection data I _2, similarly to the method of the first reconstruction of the projection data I ₁ of the first reconstruction unit 342 will be described the case of adopting the FBP method. The second reconstruction unit 343 can use the above-described reference projection data I ₀ as the reference projection data used when calculating the integral value M ₂ of the linear attenuation coefficient, but the projection data processing unit 341 performs the first projection. It is preferable to use reference projection data generated by a method similar to that used when generating the second projection data I ₂ from the data I ₁ . That is, the second reconstruction unit 343 first calculates the reference projection data I ₀ ′ from the reference projection data I _{0 according} to the following formula (3).

Then, the second reconstruction unit 343 calculates the integral value M ₂ of the linear attenuation coefficient from the second projection data I ₂ and the reference projection data I ₀ ′ according to the following equation (4).

When I ₂ (x, y) = 0, the value of the integral value M ₂ (x, y) is calculated using the upper equation of equation (4) under the same conditions as when I ₂ (x, y) ≠ 0. Becomes infinite and computation becomes impossible. Therefore, when I ₂ (x, y) = 0, the second reconstructing unit 343 replaces the second projection data I ₂ on the right side of the upper equation of Equation (4) with “1”. The integral value M ₂ (x, y) is calculated by

Next, the second reconstruction unit 343 performs one-dimensional Fourier transform on the calculated integral value M ₂ (x, y) in the channel direction. Next, the second reconstructing unit 343 performs a filtering process in the frequency direction with a high-pass filter such as a ramp filter or a Shepp-Logan filter on the value subjected to the one-dimensional Fourier transform, thereby performing a one-dimensional inverse Fourier transform. do. The second reconstruction unit 343, the data in which the one-dimensional inverse Fourier transform, all by back projection to be added for each view, the second reconstructed image S ₂ is reconstructed sectional images Generate.

  In addition, although the 1st reconstruction part 342 and the 2nd reconstruction part 343 showed all the cases where it reconfigure | reconstructs using FBP method which is an example of a back projection method, the method of reconstruction is limited to this. Instead, various reconstruction methods such as a successive approximation method may be used. For example, the successive approximation method is a method in which a pseudo provisional image is prepared in advance, and the attenuation rate is calculated by irradiating the subject with X-rays in each view. When the attenuation rate calculated for the temporary image is smaller than the measured value (attenuation rate) actually detected by the detector 13, the pixel value of the temporary image is increased. Conversely, when the attenuation rate calculated for the temporary image is larger than the measured value actually detected by the detector 13, the pixel value of the temporary image is decreased. By repeating this operation, the attenuation rate calculated in the temporary image is changed to be equal to the measured value (attenuation rate) actually detected by the detector 13 to obtain a reconstructed image. As the successive approximation method, there are an ART (Algebraic Reconstruction Technique) method, an OS-EM (Ordered Subset Expansion Maximization) method, and an ML-EM (Maximum Likelihood Exploration Method) method.

The generating unit 344 generates a restored image S _Out from the first reconstructed image S ₁ received from the first reconstructing unit 342 and the second reconstructed image S ₂ received from the second reconstructing unit 343. It is. As a first example of a reconstructed image S ₁ generated by the first reconstruction unit 342 described above, a first reconstructed image 1101 schematically illustrated in Figure 6. When the first reconstructed image S ₁ generated by the first reconstructing unit 342 includes 0 in the first projection data I ₁ , the value does not match the true linear attenuation coefficient even if a local average is taken. . In the example of the first reconstructed image 1101 shown in FIG. 6, in the cross-sectional image of the subject, the pixel value (luminance value) decreases toward the inside, and deviates from the true line attenuation coefficient. As a second example of the reconstructed image S ₂ generated by the second reconstruction unit 343, showing a second reconstructed image 1102 schematically illustrated in Figure 7. The second reconstructed image S ₂ generated by the second reconstructing unit 343 takes a local average because the number of 0 included in the second projection data I ₂ is smaller than that of the first projection data I _1. If, the image indicating a value close to the true linear attenuation coefficient than the first reconstructed image S _1. However, since the second projection data I ₂ is obtained by adding (weighted sum) pixel values, as shown in FIG. 7, the second reconstructed image S ₂ (the second reconstructed image in the example of FIG. 7). image 1102), the spatial resolution has been lower than the first reconstructed image S _1, the pattern of the cross-section of the object 40 (boundary such as an organ) is unclear.

Accordingly, generator 344, while maintaining the spatial resolution first reconstructed image S ₁ is a obtains the restored image S _Out of a pixel value close to the true linear attenuation coefficient such as the second reconstructed image S ₂ processing do. As illustrated in FIG. 2, the generation unit 344 includes a filter processing unit 3441 and a conversion unit 3442.

The filter processing unit 3441 calculates a first filter value T ₁ and a second filter value T ₂ that are local average values for each of the first reconstructed image S ₁ and the second reconstructed image S ₂ . A processing unit that performs filter processing (first filter processing and second filter processing, respectively). Here, the local average value is a value obtained by averaging pixel values in a predetermined area around the pixel of interest. For example, the local average value centered on the coordinates (m, n) in the image may be an average value in a range of ± L pixels in the m direction and the n direction with the coordinate (m, n) as the center. it can. The predetermined area may be circular.

  In addition to the simple average, the method for obtaining the local average value by the filter processing unit 3441 may be a method of taking a weighted average in which the weight is increased as the center of the predetermined region is closer. For example, an epsilon filter, a bilateral filter, or a median filter may be used.

The filter processing for calculating the first filter value T ₁ and the second filter value T ₂ are may be the same number of filter taps, or may be a different filter number of taps. For example, the second reconstructed image S _2, since the first lower spatial resolution than the reconstructed image S _1, the number of taps of the filter for calculating a second filter value T ₂ are, the first filter value T ₁ You may make it smaller than the tap number of the filter for calculating.

Further, as described above, for each of the first reconstructed image S ₁ and the second reconstructed image S _2, the first filter value T ₁ and the second filter value T ₂ of obtaining the local average value, all The pixel format may be an image format obtained by performing the filtering process, or may be a value obtained by performing the filtering process only on the pixel value to be converted by the converting unit 3442 later.

The conversion unit 3442 calculates the following equation (5) from the first reconstructed image S ₁ received from the first reconstructing unit 342 and the first filter value T ₁ and the second filter value T ₂ received from the filter processing unit 3441. ) To generate a restored image S _Out .

That is, the conversion unit 3442, as shown in equation (5), on the basis of the first filter value T ₁ and the second filter value T _2, the restored image S by converting a first reconstructed pixel values of the image S _{1 Out} is generated. Further, the qualitative significance indicated by the formula (5) will be described below. First, since the first filter value T ₁ is a local average value of the first reconstructed image S ₁ , S ₁ (m, n) / T ₁ (m, n) in the right side of Expression (5). Is approximately equal to “1” in the entire image. Therefore, the pixel value of S _Out (m, n) approximates the value of T ₂ (m, n) in the entire image, and is a pixel close to the true linear attenuation coefficient of the second reconstructed image S _2. Value. The first filter value T ₁ and the second filter value T ₂ as described above, the spatial resolution for the first is a local average value of the reconstructed image S ₁ and the second reconstructed image S ₂ each is lower Become. Therefore, the first reconstructed image S ₁ having a higher spatial resolution than the second reconstructed image S ₂ is multiplied by T ₂ (m, n) / T ₁ (m, n) as shown in Equation (5). Even so, the influence on the spatial resolution is small, and the spatial resolution of S _Out (m, n) follows the high spatial resolution of the first reconstructed image S ₁ .

Therefore, the restored image S _Out generated by the conversion unit 3442 has a pixel value close to the true linear attenuation coefficient as in the second reconstructed image S ₂ while maintaining the spatial resolution of the first reconstructed image S _1. It becomes an image. As an example of the restored image S _Out generated by the conversion unit 3442, a restored image 1201 schematically illustrated in FIG. 8 is illustrated. The restored image 1201 shown in FIG. 8 has a pixel value closer to the true linear attenuation coefficient than the first reconstructed image 1101 shown in FIG. 6, and has a higher spatial resolution than the second reconstructed image 1102 shown in FIG. It is an image.

The conversion unit 3442, based on the above equation (5), is not limited to generating a restored image S _Out, with coefficient alpha, the restored image S _Out by equation (6) below It is good also as what produces | generates.

The coefficient α in Equation (6) is a real number in the range of 0 ≦ α ≦ 1, and when α = 0, S _Out = S ₁ , and when α = 1, S _Out = (T ₂ / T ₁ ). S ₁ , that is, the same expression as Expression (5). That is, S _Out represented by Equation (6) takes a value between S ₁ and (T ₂ / T ₁ ) S ₁ based on the value of the coefficient α. For example, when the range in the channel direction in which the pixel values of the first projection data I ₁ are added (weighted sum) in the projection data processing unit 341 is increased, the spatial resolution of the second reconstructed image S ₂ is greatly reduced. for 2 reliability filter value T ₂ is reduced, it may be assumed that the smaller the coefficient alpha. In addition, when the number of zeros included in the first projection data I ₁ is small, the reliability of the first reconstructed image S ₁ is improved, so the coefficient α may be reduced.

Further, the restored image S _Out obtained by the above formula (5) or (6) may be used as an initial value of another reconstruction method. For example, an initial value such as an ML-EM method that is an example of the successive approximation method may be used. By using the restored image S _Out obtained by the above formula (5) or (6) as the initial value of the calculation of the cross-sectional image by the successive approximation method, the convergence of the image update is accelerated, and the calculation speed is increased. can do.

  Also, the projection data processing unit 341, the first reconstruction unit 342, the second reconstruction unit 343, the filter processing unit 3441, and the conversion unit 3442 in the image processing unit 34 illustrated in FIG. 2 are each realized by a program that is software. Or it may be realized by a hardware circuit. Also, the projection data processing unit 341, the first reconstruction unit 342, the second reconstruction unit 343, the generation unit 344, the filter processing unit 3441, and the conversion unit 3442 in the image processing unit 34 illustrated in FIG. However, the present invention is not limited to such a configuration.

  Further, the console device 30 of the present embodiment is configured using a computer. That is, the console device 30 includes a control device such as a CPU (Central Processing Unit) (scan control unit 33 and system control unit 36 in FIG. 1), and a storage such as a ROM (Read Only Memory) or a RAM (Random Access Memory). A device, an HDD (Hard Disk Drive), an SSD (State Solid Drive) or an external storage device such as a CD drive (image storage unit 35 in FIG. 1), and an input device such as a keyboard or mouse (input device 31 in FIG. 1) ) And a display device such as a display (display device 32 in FIG. 1). Here, as described above, at least one of the projection data processing unit 341, the first reconstruction unit 342, the second reconstruction unit 343, the filter processing unit 3441, and the conversion unit 3442 in the image processing unit 34 is a program. When implemented, the program executed by the console device 30 is an installable or executable file, such as a CD-ROM, a flexible disk (FD), a CD-R, or a DVD (Digital Versatile Disk). The program is provided by being recorded on a computer-readable recording medium. The program executed by the console device 30 may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. Further, the program executed by the console device 30 may be provided or distributed via a network such as the Internet. Further, the above program may be provided by being incorporated in advance in a ROM or the like. The program executed by the console device 30 according to the present embodiment includes the projection data processing unit 341, the first reconstruction unit 342, the second reconstruction unit 343, the filter processing unit 3441, and the conversion unit 3442 of the image processing unit 34 described above. As the actual hardware, the CPU reads and executes the program from the above-mentioned storage medium, and the above-described units are loaded and generated on the main storage device. It has become so. The above-mentioned content is the same also about the modified example mentioned later and other embodiment.

  FIG. 9 is a flowchart illustrating an example of the operation of the image processing unit according to the first embodiment. The overall operation of the image processing by the image processing unit 34 of the first embodiment will be described with reference to FIG.

<Step S11>
The projection data processing unit 341 of the image processing unit 34 receives and acquires the first projection data I ₁ that is the subject sinogram from the data collection unit 16. Then, the process proceeds to step S12.

<Step S12>
The projection data processing unit 341 calculates a weighted sum in the channel direction for the obtained pixel value of the first projection data I ₁ by the above-described equation (1), and generates second projection data I ₂ . Then, the projection data processing unit 341 sends the generated second projection data I ₂ to the second reconstruction unit 343. Then, the process proceeds to step S13.

<Step S13>
The first reconstruction unit 342 of the image processing unit 34 receives and acquires the first projection data I ₁ from the data collection unit 16. Next, the first reconstruction unit 342 calculates the integral value M ₁ of the linear attenuation coefficient from the first projection data I ₁ and the reference projection data I _{0 according} to the above equation (2). The first reconstruction unit 342 performs a one-dimensional Fourier transform on the calculated integral value M ₁ (x, y) in the channel direction by a high-pass filter such as a ramp filter or a Shepp-Logan filter. Filtering is performed in the frequency direction to perform a one-dimensional inverse Fourier transform. The first reconstruction unit 342, the data in which the one-dimensional inverse Fourier transform, by adding backprojected into each all views, the first reconstructed image S ₁ is reconstructed sectional images Generate. The first reconstruction unit 342, the first reconstructed image _{S 1} generated is sent to the filter processing section 3441 and converting section 3442 of the generator 344.

Further, the second reconstruction unit 343 of the image processing unit 34 calculates the reference projection data I ₀ ′ from the reference projection data I ₀ that is an air sinogram by the above-described equation (3). Next, the second reconstruction unit 343 calculates the integral value M ₂ of the linear attenuation coefficient from the second projection data I ₂ and the reference projection data I ₀ ′ according to the above equation (4). The second reconstruction unit 343 applies a one-dimensional Fourier transform to the calculated integral value M ₂ (x, y) in the channel direction by a high-pass filter such as a ramp filter or a Shepp-Logan filter. Filtering is performed in the frequency direction to perform a one-dimensional inverse Fourier transform. The second reconstruction unit 343, the data in which the one-dimensional inverse Fourier transform, all by back projection to be added for each view, the second reconstructed image S ₂ is reconstructed sectional images Generate. The second reconstruction unit 343, the generated second reconstructed image _{S 2,} and sends to the filter processing unit 3441. Then, the process proceeds to step S14.

<Step S14>
The filter processing unit 3441 uses a local average value for each of the first reconstructed image S ₁ received from the first reconstructing unit 342 and the second reconstructed image S ₂ received from the second reconstructing unit 343. A first filter value T ₁ and a second filter value T ₂ are calculated. The filter processing unit 3441 sends the calculated first filter value T ₁ and second filter value T ₂ to the conversion unit 3442. Then, the process proceeds to step S15.

<Step S15>
The conversion unit 3442 uses the first reconstructed image S ₁ received from the first reconstructing unit 342 and the first filter value T ₁ and the second filter value T ₂ received from the filter processing unit 3441 to obtain the above equation (5). ) Or (6) to generate a restored image S _Out . The conversion unit 3442 outputs the generated restored image S _Out to the image storage unit 35, and the image storage unit 35 stores the restored image S _Out .

  Image processing by the image processing unit 34 is executed by the operations in steps S11 to S15 described above.

As described above, the projection data processing unit 341 has fewer pixel values indicating that the number of photons is 0 than the first projection data I ₁ based on the first projection data I ₁ received from the data collection unit 16. The second projection data I ₂ is generated, and the generation unit 344 generates a restored image S _Out that is close to the pixel value of the second reconstructed image S ₂ while maintaining the spatial resolution of the first reconstructed image S _1. It is said. As a result, even under conditions in which a channel in which the number of photons detected by the detector 13 is zero is generated, a restoration composed of pixel values that are close to the true linear attenuation coefficient of the subject while suppressing a reduction in spatial resolution. A cross-sectional image that is an image can be obtained.

  The detector 13 detects the spectrum indicated by the number of photons for each energy for each channel (detection element) arranged in the circumferential direction of the rotating frame 12, but as described above, the detector 13 Detection elements are also arranged in the body axis direction of the subject 40. Therefore, a sinogram is generated for each ring-shaped array of detection elements in the body axis direction, and the above-described image processing may be executed. Alternatively, when the rotating frame 12 is continuously rotated while moving the top plate 22 as in the helical scan, not only the data detected by the same circumferential channel (detection element) is used, but the body axis direction. A sinogram may be generated by interpolating with data detected by channels that are shifted to each other. In addition, as in a dual energy X-ray CT apparatus, the energy of X-rays emitted from the X-ray tube 11 is divided into two types and switched for every round (for example, the first round is 140 [keV], 2 The circumference may be 80 [keV]), and a spectrum of different energies may be synthesized to generate a sinogram.

In addition, the pixel value of the restored image S _Out generated by the generation unit 344 is assumed to be a line attenuation coefficient, but is not limited to this, and is a value representing the amount of X-ray attenuation such as a CT value. Any are valid. Similarly, the pixel value of the sinogram is not limited to the number of photons, but is a value representing the amount of X-rays or the number of photons, or a value representing the rate of change of the amount of X-rays or the number of photons. Also good.

  Further, although the image processing until the above-described restored image is generated has been described in the X-ray inspection apparatus 1, images having different numbers of photons obtained by projecting or photographing a subject, such as an X-ray fluoroscopic apparatus, can be obtained. The present invention can be applied to all image processing apparatuses.

<Modification 1>
The image processing unit 34 according to this modification will be described focusing on differences from the image processing unit 34 according to the first embodiment. The image processing unit 34 of this modification includes a projection data processing unit 341a shown in FIG. 10 described later, instead of the projection data processing unit 341 provided in the image processing unit 34 of the first embodiment shown in FIG. .

  FIG. 10 is a diagram illustrating a block configuration of a projection data processing unit according to Modification 1 of the first embodiment. The configuration and operation of the projection data processing unit 341a included in the image processing unit 34 of this modification will be described with reference to FIG.

Projection data processing unit 341a of this modification, receives the subject sinogram which is sinogram data collection unit 16 subjects the received (see FIG. 1) 40 (see FIG. 1) as a first projection data I _1, the A processing unit that generates second projection data I ₂ in the form of a sinogram based on one projection data I ₁ . As shown in FIG. 10, the projection data processing unit 341a includes a processing unit 3411a and a determination unit 3412a.

Determination unit 3412a, based on the distribution of the pixel value indicating that the number of photons contained in the first projection data I ₁ received from the data collection unit 16 is 0, processing for generating the second projection data I ₂ A processing unit for outputting a control signal for determining a method. Here, the distribution of pixel values indicating that the number of photons is 0 is, for example, a ratio of pixel values indicating that the number of photons included in a predetermined range in the channel direction or the view direction is 0. The determination unit 3412a increases the pixel value range for calculating the weighted sum in the above equation (1) as the ratio of the pixel values indicating that the number of photons included in the predetermined range is 0 is high. Is output. For example, when the ratio of the pixel values indicating that the number of photons included in the predetermined range is 0 in the above-described formula (1) is less than 5%, the determination unit 3412a sets K = 1 to 5% or more and 10 If it is less than%, K = 5, and if it is 10% or more, a control signal such that K = 10 is output.

  The distribution of pixel values indicating that the number of photons is 0 is not limited to the ratio of pixel values indicating that the number of photons included in the predetermined range is 0. For example, the channel direction Alternatively, in the view direction or the like, the pixel value indicating that the number of photons is 0 may be the maximum width value adjacent to the pixel value.

Processing unit 3411a, based on the value of K in equation (1) is determined above by the control signal received from the determination unit 3412A, the channel direction for the first pixel value of the projection data I ₁ received from the data acquisition unit 16 or a processing unit that generates second projection data I ₂ calculates the weighted sum of such viewing direction. Processing unit 3411a sends the second projection data _{I 2} generated in the second reconstructing section 343 (see FIG. 2).

  FIG. 11 is a flowchart illustrating an example of the operation of the image processing unit according to the first modification of the first embodiment. With reference to FIG. 11, the overall operation of the image processing by the image processing unit 34 of the present modification will be described focusing on differences from the image processing by the image processing unit 34 of the first embodiment.

<< Step S11 >>
Projection data processing unit 341a of the image processing unit 34, from the data acquisition unit 16 receives and obtains the first projection data I ₁ is subject sinogram. Then, the process proceeds to step S21.

<< Step S21 >>
Based on the distribution of pixel values indicating that the number of photons included in the first projection data I ₁ received from the data collection unit 16 is 0, the determination unit 3412a of the projection data processing unit 341a performs the second projection data I _2. A control signal for determining a processing method for generating the signal is output. Then, the process proceeds to step S22.

<< Step S22 >>
Based on the processing method determined by the control signal received from the determination unit 3412a, the processing unit 3411a of the projection data processing unit 341a receives the first projection data I ₁ received from the data collection unit 16 according to the above equation (1). It calculates the weighted sum of such channel direction or view direction to generate a second projection data I ₂ for the pixel value. Processing unit 3411a sends the second projection data _{I 2} generated in the second reconstructing section 343 (see FIG. 2).

  The processes in steps S13 to S15 in FIG. 11 are the same as steps S13 to S15 in FIG. 9 of the first embodiment described above, respectively.

As described above, the projection data processor 341a on the basis of the distribution of the pixel value indicating that the number of photons contained in the first projection data I ₁ is 0, the processing for generating the second projection data I ₂ the method dynamically determines the, and generates a second projection data I ₂ on the basis of the determined processing method. Specifically, the projection data processing unit 341a, for example, based on the distribution of the pixel value indicating that the number of photons contained in the first projection data I ₁ is 0, the weighted sum in equation (1) above The range of pixel values to be calculated is dynamically determined. Thereby, compared with the first embodiment, more suitable second projection data I ₂ is generated, that is, the second projection data I ₂ with high reliability with a small pixel value indicating that the number of photons is 0. Can be generated. Therefore, a cross-sectional image that is a restored image composed of pixel values close to the true linear attenuation coefficient of the subject can be obtained with higher accuracy.

  Note that the processing unit 3411a, the determination unit 3412a, the first reconstruction unit 342, the second reconstruction unit 343, the filter processing unit 3441, and the conversion unit 3442 in the image processing unit 34 of the present modification are each a program that is software. It may be realized or may be realized by a hardware circuit. Further, the projection data processing unit 341a shown in FIG. 10 conceptually shows functions, and is not limited to such a configuration.

<Modification 2>
The image processing unit 34 according to this modification will be described focusing on differences from the image processing unit 34 according to the first embodiment. The image processing unit 34 of this modification includes a projection data processing unit 341b shown in FIG. 12 described later, instead of the projection data processing unit 341 provided in the image processing unit 34 of the first embodiment shown in FIG. .

  FIG. 12 is a diagram illustrating a block configuration of a projection data processing unit according to Modification 2 of the first embodiment. The configuration and operation of the projection data processing unit 341b included in the image processing unit 34 of the present modification will be described with reference to FIG.

Projection data processor 341b of this modified example receives the subject sinogram which is sinogram data collection unit 16 subjects the received (see FIG. 1) 40 (see FIG. 1) as a first projection data I _1, the A processing unit that generates second projection data I ₂ in the form of a sinogram based on one projection data I ₁ . As shown in FIG. 12, the projection data processing unit 341b includes a processing unit 3411b and an analysis unit 3412b.

Processing unit 3411b, depending the above equation (1), the second projection data I ₂ calculates the weighted sum of the channel direction or view direction, etc. for the first pixel value of the projection data I ₁ received from the data acquisition unit 16 A processing unit to be generated. Processing unit 3411b sends the second projection data _{I 2} generated in the second reconstructing section 343 (see FIG. 2) or the analysis unit 3412B.

Analyzer 3412b, based on the distribution of the pixel value indicating that the number of photons in the second projection data _{I 2} received from the processing unit 3411b is 0, the processing unit 3411b of the second generation of projection data _{I 2} is It is a processing unit that analyzes whether it is necessary again and outputs a control signal including the analysis result. For example, as a distribution of pixel values indicating that the number of photons is 0, the analysis unit 3412b has a predetermined ratio of pixel values indicating that the number of photons included in a predetermined range in the channel direction or the view direction is 0. If it is higher, it is determined that the generation of the second projection data I ₂ is necessary again, and a control signal instructing the generation of the second projection data I ₂ is output.

Processing unit 3411b, when receiving the control signal for instructing the generation again from analysis section 3412B, as compared with the case previously generated, the second projection data I _2, calculates a weighted sum in equation (1) above and wide range of pixel values, again, to generate a second projection data I _2. For example, the processing unit 3411b may last a value of K of the formula (1) described above in the case of generating the second projection data I ₂ and doubled again, to generate a second projection data I _2.

  When the processing unit 3411b receives a control signal instructing generation again from the analysis unit 3412b, the processing unit 3411b previously calculated a weighted sum in a predetermined range in the channel direction in the above equation (1). The weighted sum may be calculated in the direction, slice direction, time direction, energy direction, or the like.

  FIG. 13 is a flowchart illustrating an example of the operation of the image processing unit according to the second modification of the first embodiment. With reference to FIG. 13, the overall operation of the image processing by the image processing unit 34 of the present modification will be described focusing on differences from the image processing by the image processing unit 34 of the first embodiment.

<< Step S11 >>
Projection data processing unit 341b in the image processing unit 34, from the data acquisition unit 16 receives and obtains the first projection data I ₁ is subject sinogram. Then, the process proceeds to step S31.

<< Step S31 >>
Processing unit 3411b of the projection data processor 341b, due the above equation (1), the first pixel value of the projection data I ₁ received from the data acquisition unit 16 calculates a weighted sum, such as the channel direction or view direction, the second generates the projection data _{I 2.} Then, the processing section 3411b the generated second projection data _{I 2,} and sends the analysis unit 3412B. Then, the process proceeds to step S32.

<< Step S32 >>
Analyzer 3412b, based on the distribution of the pixel value indicating that the number of photons in the second projection data _{I 2} received from the processing unit 3411b is 0, the processing unit 3411b of the second generation of projection data _{I 2} is It analyzes whether it is necessary again and outputs a control signal including the analysis result. For example, as a distribution of pixel values indicating that the number of photons is 0, the analysis unit 3412b has a predetermined ratio of pixel values indicating that the number of photons included in a predetermined range in the channel direction or the view direction is 0. If it is higher, it is determined that the generation of the second projection data I ₂ is necessary again, and a control signal instructing the generation of the second projection data I ₂ is output. Then, the process proceeds to step S33.

<< Step S33 >>
Processing unit 3411b, when the analysis section 3412B, receiving the control signal for instructing the generation again (step S33: Yes), the last time, as compared with the case of generating the second projection data _{I 2,} the above-mentioned formula ( In 1), the range of pixel values for calculating the weighted sum is widened. Then, the process returns to step S31. Processing unit 3411b, when the analysis section 3412B, receiving the control signal indicating that there is no need of re-generation (step S33: No), the second projection data _{I 2} second reconstruction unit 343 generates (FIG. 2 See). Then, the process proceeds to step S13.

  The processes in steps S13 to S15 in FIG. 13 are the same as steps S13 to S15 in FIG. 9 of the first embodiment described above, respectively.

As described above, the projection data processing unit 341b, based on the distribution of the pixel value indicating that the number of photons in the second projection data I ₂ generated from first projection data I ₁ is 0, the second projection generation of the data I ₂ analyzes whether needed again. Projection data processor 341b, as a result of the analysis, when the second generation of projection data I ₂ is determined to be needed again, and a wide range of pixel values to calculate a weighted sum in equation (1) above again, to generate a second projection data _{I 2.} Thereby, compared with the first embodiment, more suitable second projection data I ₂ is generated, that is, the second projection data I ₂ with high reliability with a small pixel value indicating that the number of photons is 0. Can be generated. Therefore, a cross-sectional image that is a restored image composed of pixel values close to the true linear attenuation coefficient of the subject can be obtained with higher accuracy.

  Note that the processing unit 3411b, the analysis unit 3412b, the first reconstruction unit 342, the second reconstruction unit 343, the filter processing unit 3441, and the conversion unit 3442 in the image processing unit 34 of the present modification are programs that are software. It may be realized or may be realized by a hardware circuit. Further, the projection data processing unit 341b shown in FIG. 12 conceptually shows functions, and is not limited to such a configuration.

(Second Embodiment)
The X-ray inspection apparatus according to the second embodiment will be described focusing on differences from the X-ray inspection apparatus 1 according to the first embodiment. The X-ray inspection apparatus according to the second embodiment has the same configuration as the X-ray inspection apparatus 1 according to the first embodiment shown in FIG. 1, but instead of the image processing unit 34, FIG. An image processing unit 34a is provided. In the present embodiment, according to the reliability of the first projection data I _1, an operation for switching the range for calculating the weighted sum in the above formula (1) when generating the second projection data I ₂ , and operation will be described for switching whether to perform the conversion process of the first reconstructed image S ₁ of the pixel value in the conversion unit 3442.

  FIG. 14 is a diagram illustrating an example of a block configuration of an image processing unit according to the second embodiment. With reference to FIG. 14, an outline of the configuration and operation of the image processing unit 34a of the second embodiment will be described.

  As illustrated in FIG. 14, the image processing unit 34 a includes a projection data processing unit 341, a first reconstruction unit 342, a second reconstruction unit 343, a generation unit 344, and a reliability calculation unit 345. ing.

The projection data processing unit 341 receives the subject sinogram as the first projection data I ₁ from the data collection unit 16, receives first selection information described later from the system control unit 36, and receives the first projection data I _1. And a processing unit that generates the second projection data I ₂ based on the first selection information. Specifically, the projection data processing unit 341 calculates a weighted sum in the channel direction for the pixel value of the first projection data I ₁ as the above-described formula (1), and sets it as the second projection data I ₂ . . At this time, the projection data processing unit 341 calculates the weighted sum according to the range of pixel values for calculating the weighted sum indicated by the first selection information.

In the above formula (1), the weighted sum of the first projection data I _{1 in} the channel direction is calculated to be the second projection data I _2. However, the present invention is not limited to this. as mentioned in the embodiment, first projection data I ₁ of the view direction, slice direction may be a second projection data I ₂ calculates the weighted sum of the time direction or energy direction.

  The operations of the first reconstruction unit 342 and the second reconstruction unit 343 are the same as the operations of the first reconstruction unit 342 and the second reconstruction unit 343 of the image processing unit 34 of the first embodiment, respectively.

The generating unit 344 generates a restored image S _Out from the first reconstructed image S ₁ received from the first reconstructing unit 342 and the second reconstructed image S ₂ received from the second reconstructing unit 343. It is. As illustrated in FIG. 14, the generation unit 344 includes a filter processing unit 3441 and a conversion unit 3442.

  The filter processing unit 3441 is the same as the operation of the filter processing unit 3441 of the image processing unit 34 of the first embodiment.

The conversion unit 3442 uses the first reconstructed image S ₁ received from the first reconstructing unit 342 and the first filter value T ₁ and the second filter value T ₂ received from the filter processing unit 3441 to obtain the above equation (5). ) To generate a restored image S _Out . That is, the conversion unit 3442 converts and restores the pixel value of the first reconstructed image S ₁ based on the first filter value T ₁ and the second filter value T ₂ as shown in the above equation (5). An image S _Out is generated. The conversion unit 3442 receives the second selection information described later from the system control unit 36, the second selection information performs conversion processing of the first reconstructed image S ₁ of the pixel values indicated by the above formula (5) If it indicates, the conversion process is performed. On the other hand, the conversion unit 3442, a second selection information may indicate that that does not perform conversion processing of the first reconstructed image S ₁ of the pixel values indicated by the above equation (5), received from the first reconstructing section 342 It outputs the first reconstructed image _{S 1} as it is as a restored image _{S Out.} The conversion unit 3442 transmits the generated restored image S _Out to the system control unit 36.

The reliability calculation unit 345 receives the first projection data I ₁ from the data collection unit 16 (see FIG. 1), receives the second projection data I ₂ from the projection data processing unit 341, and receives the first projection data I ₁ and a processing unit for calculating a second confidence of projection data I _2. Here, the calculated reliability is calculated based on a distribution of pixel values indicating that the number of photons included in each of the first projection data I ₁ and the second projection data I ₂ is 0, and the number of photons is by including a pixel value indicating a 0, is expected to linear attenuation coefficient is first reconstructed image S ₁ and the second reconstructed pixel values of the image S ₂ reconstructed departing from the true value, respectively This is a lower value. The reliability is, for example, a value in the range of 0 to 100%. Further, the distribution of pixel values indicating that the number of photons is 0 is as described in the first embodiment.

The input device 31 is a device for an operator who operates the X-ray inspection apparatus according to the present embodiment to input various instructions, and is a device that transmits various commands input to the system control unit 36. Further, the input device 31 receives an operation input for switching the extent of the range for calculating the weighted sum in the above-described equation (1) when generating the second projection data I _2, and receives the operation signal as the system control unit 36. Send to. Further, the input device 31 receives an operation input for switching whether or not to perform the conversion processing of the pixel value of the first reconstructed image S ₁ in the conversion unit 3442, and transmits the operation signal to the system control unit 36.

  The display device 32 displays a GUI (Graphical User Interface) for accepting an operation instruction from the operator via the input device 31, or displays a reconstructed image (restored image, cross-sectional image) stored in the image storage unit 35. It is a device to do. The display device 32 is image-processed by the image processing unit 34a and calculated by the reconstructed image (restored image, cross-sectional image) received via the system control unit 36, and the reliability calculation unit 345, and the system control unit The reliability received via 36 is displayed.

The system control unit 36 is a processing unit that controls the entire X-ray inspection apparatus by controlling the operations of the gantry device 10, the couch device 20, and the console device 30. In addition, the system control unit 36 receives the reliability from the reliability calculation unit 345, receives the restored image from the conversion unit 3442, and transmits the reliability and the restored image to the display device 32. Further, the system control unit 36 generates first selection information indicating a range of pixel values for calculating the weighted sum in the above-described equation (1) by the projection data processing unit 341 based on the operation signal received from the input device 31. Then, the first selection information is transmitted to the projection data processing unit 341. Further, the system control unit 36 indicates whether or not to perform the pixel value conversion processing of the first reconstructed image S ₁ represented by the above-described equation (5) based on the operation signal received from the input device 31. The selection information is generated, and the second selection information is transmitted to the conversion unit 3442.

  Further, the projection data processing unit 341, the first reconstruction unit 342, the second reconstruction unit 343, the filter processing unit 3441, the conversion unit 3442, and the reliability calculation unit 345 in the image processing unit 34a illustrated in FIG. It may be realized by a program or a hardware circuit. In addition, the projection data processing unit 341, the first reconstruction unit 342, the second reconstruction unit 343, the generation unit 344, the filter processing unit 3441, the conversion unit 3442, and the reliability calculation unit 345 in the image processing unit 34a illustrated in FIG. Is a conceptual illustration of the function and is not limited to such a configuration.

  FIG. 15 is a diagram illustrating an example of display contents of the display device according to the second embodiment. The display contents displayed on the display device 32 and the selection operation by the input device 31 will be described with reference to FIG.

  As illustrated in FIG. 15, the display device 32 includes a presentation unit 321, a selection unit 322, and an image display unit 323.

The presentation unit 321 is an area that displays the reliability of the first projection data I ₁ and the second projection data I ₂ calculated by the reliability calculation unit 345. In the example of FIG. 15, “the reliability of the first image” corresponds to the reliability of the first projection data I ₁ and is displayed as “60%”, and “the reliability of the second image” is the second projection data. This corresponds to the reliability of I ₂ and is displayed as “95%”.

Selecting unit 322, the conversion unit 3442, the display whether to perform the conversion process of the first reconstructed image S ₁ of the pixel values indicated by the above equation (5), and conversion processing ON button for selecting operations 3221a and a conversion processing OFF button 3221b. For example, when the user selects to perform the conversion processing of the first reconstructed image S ₁ of pixel values, it performs a click operation by the input device 31 such as a mouse conversion processing ON button 3221a. Thus, as shown in FIG. 15, the conversion process ON button 3221a is highlighted in black, it appears that it has been selected to perform the conversion process of the first reconstructed image S ₁ of the pixel values.

  In addition, the selection unit 322 displays the degree of the pixel value range for calculating the weighted sum in the above-described equation (1) by the projection data processing unit 341, and performs the selection operation with a weak button 3222a, a middle button 3222b, and a strong button. A button 3222c is provided. For example, when the user selects an intermediate range of pixel values for calculating the weighted sum, the user clicks the middle button 3222b with the input device 31 such as a mouse. As a result, as shown in FIG. 15, the middle button 3222b is highlighted in black, indicating that the range of pixel values for calculating the weighted sum has been selected to be a medium range. Also, for example, when the weak button 3222a is clicked, K = 1 in the above formula (1), when the middle button 3222b is clicked, K = 5, and when the strong button 3222c is clicked , K = 10.

  The display device 32 and the input device 31 may be realized by, for example, an integrated touch panel. In this case, the conversion processing ON button 3221a, the conversion processing OFF button 3221b, the weak button 3222a, the middle button 3222b, and the strong button 3222c may be a touch operation that can be directly touched by the user instead of a click operation using a mouse or the like.

The image display unit 323 is an area for displaying the restored image S _Out generated by the conversion unit 3442. The image display unit 323 displays the first reconstructed image S ₁ generated by the first reconstructing unit 342 or the second reconstructed image S ₂ generated by the second reconstructing unit 343. Alternatively, the first reconstructed image S ₁ , the second reconstructed image S ₂ , and the restored image S _Out may all be displayed by dividing the area.

  As shown in FIG. 15, the display device 32 includes the presentation unit 321, the selection unit 322, and the image display unit 323 on the same display screen, but is not limited thereto. For example, at least one of the presentation unit 321, the selection unit 322, and the image display unit 323 may be displayed on a separate display device, or all may be displayed on different display devices.

  FIG. 16 is a flowchart illustrating an example of the operation of the image processing unit according to the second embodiment. The overall operation of the image processing by the image processing unit 34a of the second embodiment will be described with reference to FIG.

<Step S41>
The projection data processing unit 341 of the image processing unit 34 a receives and acquires the first projection data I ₁ that is the subject sinogram from the data collection unit 16. Then, the process proceeds to step S42.

<Step S42>
The projection data processing unit 341 receives the subject sinogram as the first projection data I ₁ from the data collection unit 16 and receives the first selection information from the system control unit 36, and receives the first projection data I ₁ and the first projection data I ₁ . Based on the one selection information, the second projection data I ₂ is generated. Specifically, the projection data processing unit 341 calculates the pixel value of the first projection data I ₁ according to the range of pixel values for calculating the weighted sum indicated by the first selection information, as shown in the above equation (1). for the second projection data I ₂ calculates the weighted sum of the channel direction. Then, the projection data processing unit 341 sends the generated second projection data I ₂ to the second reconstruction unit 343 and the reliability calculation unit 345. Then, the process proceeds to step S43.

<Steps S43 and S44>
The processes in steps S43 and S44 are the same as the processes in steps S13 and S14 shown in FIG. 9, respectively. Then, the process proceeds to step S45.

<Step S45>
Reliability calculation unit 345 of the image processing section 34a receives the first projection data _{I 1} from the data acquisition unit 16 (see FIG. 1), receives the second projection data _{I 2} from the projection data processing unit 341, first It calculates the reliability of the projection data I ₁ and the second projection data I _2. Then, the reliability calculation unit 345 transmits the calculated reliability to the system control unit 36. Then, the process proceeds to step S46.

<Step S46>
The system control unit 36 transmits the reliability of the first projection data I ₁ and the second projection data I ₂ received from the reliability calculation unit 345 to the display device 32. Display device 32, a first reliability of projection data I ₁ and the second projection data I ₂ received and presented to the user by displaying the presentation unit 321. Then, the process proceeds to step S47.

<Step S47>
The user confirms the reliability of the first projection data I ₁ and the second projection data I ₂ presented on the presentation unit 321 of the display device 32, and the _first equation (5) described above with respect to the conversion unit 3442 indicates it is determined whether to perform the conversion process of the pixel value of the reconstructed image S _1, to the conversion processing ON button 3221a or conversion OFF button 3221b selection operation (click operation using a mouse or the like as an input device 31). The input device 31 receives an operation input to the conversion process ON button 3221 a or the conversion process OFF button 3221 b by the user, and transmits the operation signal to the system control unit 36. The system control unit 36 based on the operation signal received from the input device 31, a second selection information indicating whether to perform conversion processing of the first reconstructed image S ₁ of the pixel values indicated by the above formula (5) And the second selection information is transmitted to the conversion unit 3442. Then, the process proceeds to step S48.

<Step S48>
Conversion unit 3442, determines whether a second selection information received from the system controller 36, indicates either whether or not to perform the conversion process of the first reconstructed image S ₁ of the pixel values indicated by the above formula (5) To do. When the second selection information indicates that the conversion process is to be performed (step S48: Yes), the process proceeds to step S49. On the other hand, when the second selection information is shown not to perform conversion processing (step S48: No), the conversion unit 3442, as the first reconstructed image _{S 1} received from the first reconstructing section 342 A restored image S _Out is generated, and the restored image S _Out is transmitted to the display device 32. Then, the process proceeds to step S50.

<Step S49>
The conversion unit 3442 uses the first reconstructed image S ₁ received from the first reconstructing unit 342 and the first filter value T ₁ and the second filter value T ₂ received from the filter processing unit 3441 to obtain the above equation (5). ) To convert the pixel value of the _first reconstructed image S1 to generate a restored image _SOut . The conversion unit 3442 transmits the generated restored image S _Out to the display device 32. Then, the process proceeds to step S50.

<Step S50>
The display device 32 displays the received restored image S _Out on the image display unit 323.

Image processing by the image processing unit 34a is executed by the operations in steps S41 to S50 described above. The user can also perform the selection operation again in the selection unit 322 by confirming the restored image S _Out displayed on the image display unit 323 of the display device 32.

As described above, in the present embodiment, while the user confirms the first projection data I ₁ and the second projection data T ₂ of the reliability or the restored image S _Out, above equation by the projection data processor 341 (1 selection operation range of pixel values to calculate a weighted sum in), and, selecting whether to perform the conversion process of the first reconstructed image S ₁ of the pixel values indicated by the above equation by the conversion section 3442 (5) operation can do. This makes it possible to accurately obtain a cross-sectional image, which is a restored image composed of pixel values close to the true linear attenuation coefficient of the subject, while effectively suppressing the reduction in spatial resolution based on the determination by the user. it can.

Note that above at the timing of step S46, the user confirms the first reliability of projection data _{I 1} and the second projection data _{I 2} presented by the presentation unit 321 of the display device 32, the converted part 3442 of formula (5) it is assumed that the selection operation whether to perform the first conversion processing pixel values of the reconstructed image S ₁ indicated, but is not limited thereto. That is, instead of prompting the user to select whether or not to perform the conversion process at the timing of step S46, the conversion unit according to the content that has been selected and operated in advance before the image processing operation by the image processing unit 34a. 3442 may be as to determine whether to perform the conversion process of the first reconstructed image S ₁ of the pixel values indicated by the above equation (5). In this case, the reliability of the first projection data I ₁ and the second projection data I ₂ finally obtained or the restored image S _Out displayed on the image display unit 323 is confirmed, and the first reconstruction is performed. whether to perform the conversion process of the pixel values of the image S ₁ may be assumed that the selection operation again. Or when the selection operation which does not perform the conversion process is performed on the content selected before the image processing operation by the image processing unit 34a, the processing in steps S42 and S44 and the second reconstructing unit 343 in step S43 are performed. the second process of generating reconstructed image S ₂ due can be skipped.

In step S42, the projection data processing unit 341 performs the first calculation according to the above equation (1) according to the range of pixel values for calculating the weighted sum indicated by the first selection information generated in advance by the system control unit 36. Although the pixel values of the projection data I ₁ are assumed for calculating the weighted sum of the channel direction, but is not limited thereto. That is, in step S42, before the projection data processing unit 341 generates the second projection data I _2, to the user, the pixel values for calculating the weighted sum in the above equation by the projection data processor 341 (1) The selection unit 322 may be prompted to perform a selection operation on the range. In this case, the user refers to the reliability of the first projection data I ₁ and the second projection data I ₂ previously presented by the presentation unit 321 or the restored image S _Out displayed on the image display unit 323. The selection operation may be performed. Further, in step S45, the reliability calculation unit 345, although assumed to calculate the reliability of the first projection data I _1, before generating the projection data processing unit 341 according to the second projection data I ₂ in step S42 prior to, the reliability calculation unit 345 calculates a first reliability of projection data I _1, the presentation unit 321 may be configured to be presented first reliability of projection data I _1.

In the second embodiment, the pixel value of the first reconstructed image S ₁ represented by the above-described equation (5) is converted by the conversion unit 3442 on the screen of the display device 32 shown in FIG. The selection operation of whether or not to perform the process and the selection operation of the pixel value range for calculating the weighted sum in the above-described formula (1) by the projection data processing unit 341 is performed, but the present invention is not limited to this. It is not a thing. For example, the reliability calculation unit 345 calculates the reliability of the first projection data I ₁ prior to the generation of the second projection data I ₂ by the projection data processing unit 341 in step S42. Based on the reliability of one projection data I ₁ , the projection data processing unit 341 determines a range of pixel values for calculating the weighted sum in the above-described equation (1), and the conversion unit 3442 includes the above-described equation (5). it may alternatively be determined whether to perform the first conversion processing pixel values of the reconstructed image S ₁ indicated.

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

DESCRIPTION OF SYMBOLS 1 X-ray inspection apparatus 10 Mount apparatus 11 X-ray tube 11a X-ray beam 12 Rotating frame 13 Detector 14 Irradiation control part 15 Mount drive part 16 Data collection part 20 Bed apparatus 21 Bed drive apparatus 22 Top plate 30 Console apparatus 31 Input apparatus 32 Display device 33 Scan control unit 34, 34a Image processing unit 35 Image storage unit 36 System control unit 40 Subject 41 Projected section 321 Presentation unit 322 Selection unit 323 Image display unit 341, 341a, 341b Projection data processing unit 342 First re- Configuration unit 343 Second reconstruction unit 344 Generation unit 345 Reliability calculation unit 1001 Sinogram 1011, 1011a to 1011d Subject sinogram 1101 First reconstruction image 1102 Second reconstruction image 1201 Restored image 3221a Conversion processing ON button 3221b Conversion processing FF button 3222a weak button 3222b in button 3222c little buttons 3411a, 3411b processor 3412a determining unit 3412b analyzer 3441 filter processor 3442 converting unit

Claims (14)

An image processing apparatus that performs image processing on an image based on energy of radiation that has passed through a subject,
Projection for generating second projection data having a pixel value as a value obtained by a predetermined calculation based on a pixel value in a predetermined range in a predetermined direction of the first projection data based on photons of the energy of the radiation transmitted through the subject. A data processing unit;
A first reconstruction unit that reconstructs the first projection data to generate a first reconstructed image;
A second reconstruction unit that reconstructs the second projection data to generate a second reconstructed image;
A generating unit that generates a restored image by converting the first reconstructed image based on the second reconstructed image;
An image processing apparatus.   The projection data processing unit performs the predetermined calculation on the first projection data to obtain the second projection data in which the pixel value indicating that the number of photons is 0 is smaller than the first projection data. The image processing apparatus according to claim 1, which is generated. The generator is
A filter processing unit that calculates a first filter value from the first reconstructed image by a first filter process and calculates a second filter value from the second reconstructed image by a second filter process;
A conversion unit that converts the first reconstructed image to generate the restored image based on the first filter value and the second filter value;
The image processing apparatus according to claim 1, further comprising:   The image processing device according to claim 3, wherein the conversion unit converts the first reconstructed image to generate the restored image based on a value of a ratio between the second filter value and the first filter value. .   The first filter process and the second filter process are a filter process for calculating an average value, a filter process for calculating a weighted average, a filter process using an epsilon filter, a filter process using a bilateral filter, or a filter process using a median filter. The image processing apparatus according to claim 3, wherein the image processing apparatus is any one of the processes.   The image processing apparatus according to claim 1, wherein the predetermined direction is any one of a channel direction, a view direction, a slice direction, a time direction, or an energy direction of the first projection data, or a combination of two or more. .   The predetermined calculation is one of an operation for obtaining a weighted sum for the pixel values in the predetermined range, a calculation by a non-linear filter for the pixel values in the predetermined range, or an operation for subsampling from the weighted sum to obtain a pixel value. The image processing apparatus according to claim 1 or 2. The projection data processing unit
A determination unit that determines the content of the predetermined calculation for generating the second projection data based on a distribution of pixel values of 0 included in the first projection data;
A processing unit that generates the second projection data from the first projection data by the predetermined calculation of the content determined by the determination unit;
The image processing apparatus according to claim 1, further comprising: The projection data processing unit
A processing unit that generates the second projection data from the first projection data by the predetermined calculation;
An analysis unit that analyzes whether generation of the second projection data is necessary again based on a distribution of 0 pixel values included in the second projection data;
Have
When the analysis result of the analysis unit indicates that the second projection data needs to be generated again, the processing unit changes the content of the predetermined calculation and generates the second projection data again. The image processing apparatus according to claim 1 or 2. A reliability calculation unit that calculates the reliability of the first projection data based on a distribution of 0 pixel values included in the first projection data;
The conversion unit generates the restored image by converting the first reconstructed image based on the first filter value and the second filter value based on the reliability, or the first reconstructed image The image processing apparatus according to claim 3, wherein it is determined whether the restored image is directly used as the restored image. A reliability calculation unit that calculates the reliability of the first projection data based on a distribution of 0 pixel values included in the first projection data;
The image processing apparatus according to claim 1, wherein the projection data processing unit determines the content of the predetermined calculation for generating the second projection data based on the reliability. A reliability calculation unit that calculates the reliability of the first projection data based on a distribution of 0 pixel values included in the first projection data;
A presentation unit for presenting the reliability;
Whether the converting unit converts the first reconstructed image based on the first filter value and the second filter value to generate the restored image, or uses the first reconstructed image as it is as the restored image Or a selection unit that receives an operation input for selecting the content of the predetermined calculation for the projection data processing unit to generate the second projection data;
The image processing apparatus according to any one of claims 3 to 5, further comprising: The image processing apparatus according to any one of claims 1 to 12,
An X-ray tube that irradiates the radiation around the subject;
A detector for detecting the energy of the radiation emitted from the X-ray tube;
X-ray CT apparatus provided with A program for executing image processing on an image based on the energy of the radiation that has passed through a subject irradiated with radiation,
Projection for generating second projection data having a pixel value as a value obtained by a predetermined calculation based on a pixel value in a predetermined range in a predetermined direction of the first projection data based on photons of the energy of the radiation transmitted through the subject. Data processing steps;
A first reconstruction step of reconstructing the first projection data to generate a first reconstructed image;
A second reconstruction step of reconstructing the second projection data to generate a second reconstructed image;
Generating a restored image by converting the first reconstructed image based on the second reconstructed image;
A program that causes a computer to execute.