Pcct device

Abstract

The present application provides a PCCT device that can correct the band artifacts of one material discrimination image together with the band artifacts of another material discrimination image. The PCCT device acquires projection data divided into a plurality of energy bins by irradiating X-rays to an object, and is characterized by comprising: a first correction unit that corrects the band artifacts of a first material discrimination image among a plurality of material discrimination images created based on the projection data, and calculates a correction amount of the band artifacts, i.e., a first correction amount; an energy calculation unit that calculates an average energy of X-rays that have passed through the object; and a second correction unit that corrects the band artifacts of a second material discrimination image using a correction amount, i.e., a second correction amount, calculated based on the first correction amount and the average energy.

Description

Technical Field

[0001] The present invention relates to the correction of artifacts in a substance identification image, wherein the substance identification image is based on projection data divided into multiple energy bins obtained by a photon counting detector or the like, and the substance in the photographed body is identified. Background Technology

[0002] PCCT (Photon Counting Computed Tomography) devices equipped with detectors that employ photon counting, i.e., photon-counting detectors, can display medical images containing more information than existing CT devices. For example, they can display images made using projection data divided into multiple energy chambers, i.e., energy chamber images, and images that distinguish multiple substances, i.e., substance-discrimination images.

[0003] However, in X-ray CT images, beam hardening artifacts can sometimes occur in areas with high X-ray absorption coefficients, such as bone tissue and therapeutic metal components within the body. These artifacts can manifest as dark bands with lower CT values ​​between areas of high X-ray absorption. Beam hardening artifacts can be corrected, for example, by methods disclosed in Patent Document 1.

[0004] Existing technical documents

[0005] Patent documents

[0006] Patent Document 1: JP 2009-50413

[0007] However, Patent Document 1 does not consider banded artifacts such as dark bands generated in the substance identification image. When a dark band is generated in one substance identification image, a bright band is generated in another substance identification image; simply correcting one banded artifact is insufficient. Summary of the Invention

[0008] Therefore, the object of the present invention is to provide a PCCT device that can correct band artifacts in one material discrimination image together with band artifacts in another material discrimination image.

[0009] To achieve the above objectives, the present invention provides a PCCT device that acquires projection data divided into multiple energy chambers by irradiating a subject with X-rays. The PCCT device is characterized by comprising: a first correction unit that corrects banded artifacts in a first material discrimination image among multiple material discrimination images generated based on the projection data, and calculates a correction amount for the banded artifacts, i.e., a first correction amount; an energy calculation unit that calculates the average energy of the X-rays transmitted through the subject; and a second correction unit that uses a correction amount calculated based on the first correction amount and the average energy, i.e., a second correction amount, to correct banded artifacts in a second material discrimination image.

[0010] Invention Effects

[0011] According to the present invention, a PCCT device is provided that can correct band artifacts in one material discrimination image together with band artifacts in another material discrimination image. Attached Figure Description

[0012] Figure 1 This is a diagram showing the overall structure of a PCCT device.

[0013] Figure 2 This is a diagram illustrating an example of X-rays divided into multiple energy chambers.

[0014] Figure 3 It is a graph illustrating the data for substance identification.

[0015] Figure 4 This is a diagram representing an example of a substance discrimination mapping.

[0016] Figure 5 This is a diagram illustrating the banded artifacts in a substance identification image.

[0017] Figure 6 This is a diagram showing the functional blocks of Embodiment 1.

[0018] Figure 7 This is a diagram illustrating an example of the processing flow of Example 1.

[0019] Figure 8 This is a diagram illustrating an example of the correction process for a substance identification image.

[0020] Explanation of reference numerals in the attached figures

[0021] 100: PCCT unit, 101: Subject, 102: Stage, 200: Input / output unit, 210: Input device, 220: Monitor, 300: Imaging unit, 310: X-ray source, 311: Collimator, 320: X-ray detector, 321: Detection element, 330: Stand, 331: Opening, 332: Rotating plate, 340: Imaging control unit, 341: X-ray control unit, 342: Stand control unit, 343: Stage control unit, 344: Detector control unit, 400: Overall control unit, 4 01: CPU; 402: Memory; 403: Storage device; 501: X-ray focus; 510: Correction component; 511: First substrate material; 512: Second substrate material; 520: Material identification data; 521: Material identification mapping; 531: Aluminum; 532: Acrylic resin; 540: Aluminum image; 541: Aluminum region; 542: Dark band; 550: Acrylic resin image; 551: Acrylic resin region; 552: Bright band; 601: First correction unit; 602: Energy calculation unit; 603: Second correction unit. Detailed Implementation

[0022] The following description, with reference to the accompanying drawings, illustrates embodiments of the PCCT (Photon Counting Computed Tomography) device according to the present invention. Furthermore, in the following description and drawings, components having the same functional structure are labeled with the same reference numerals to avoid redundant descriptions.

[0023]

Example 1

[0024] use Figure 1 The overall structure of the PCCT device 100 will be described below. The PCCT device 100 includes an input / output unit 200, an imaging unit 300, and an overall control unit 400.

[0025] The input / output unit 200 includes an input device 210 and a monitor 220. The input device 210 is a device used by the operator to input shooting conditions, etc., such as a mouse or keyboard. The monitor 220 is a display device for outputting the input shooting conditions, etc., and in the case of having a touch panel function, it also serves as the input device 210.

[0026] The imaging unit 300 includes an X-ray source 310, an X-ray detector 320, a gantry 330, a worktable 102, and an imaging control unit 340 to acquire projection data of the subject 101 at various projection angles. Furthermore, the acquired projection data is divided into multiple energy chambers.

[0027] X-ray source 310 is a device for irradiating the subject 101 with X-rays. A collimator 311 is provided between the X-ray source 310 and the subject 101. The collimator 311 is a device for adjusting the length of the X-rays irradiating the subject 101 in the z-direction.

[0028] The X-ray detector 320 is a device for detecting X-rays, i.e., direct rays, that pass through the subject 101 without scattering, and has multiple detection elements 321. Approximately 1000 detection elements 321 are arranged at equal distances, for example, 1000 mm, from the X-ray generation point of the X-ray source 310. Each detection element 321 is an element that detects X-rays and outputs an electrical signal corresponding to the amount of X-rays incident on it. The detection elements 321 are arranged in the xy plane and have a size of, for example, 0.5 mm square. The detection elements 321 can be indirect detection elements combining scintillator elements and photodiode elements, or they can be semiconductor detection elements represented by CdTe. In indirect detection elements, the scintillator element produces fluorescence due to the incident X-rays, and this fluorescence is converted into an electrical signal in the photodiode element.

[0029] Detection element 321 detects incident X-rays as... Figure 2 As shown, it is divided into multiple energy chambers for testing. Figure 2 In the diagram, the X-rays to be detected, divided into three energy chambers—T1-T2, T2-T3, and T3-—are represented as bin1, bin2, and bin3. Furthermore, the X-rays from various combinations of energy chambers can be calculated from the X-rays detected in multiple energy chambers. For example, the X-ray intensities of bin1+bin2 and bin2+bin3 can be calculated from the X-ray intensities of bin1, bin2, and bin3.

[0030] Back Figure 1 The frame 330 has a circular opening 331 at its center for arranging a stage 102 for mounting the subject 101. The diameter of the opening 331 is, for example, 700 mm. A rotating plate 332 is provided within the frame 330, which mounts an X-ray source 310 and an X-ray detector 320, allowing the X-ray source 310 and X-ray detector 320 to rotate around the subject 101. The stage 102 moves in the z-direction to adjust the position of the subject 101 relative to the frame 330.

[0031] The imaging control unit 340 includes an X-ray control unit 341, a gantry control unit 342, a stage control unit 343, and a detector control unit 344. The X-ray control unit 341 controls the voltage applied to the X-ray source 310, etc. The gantry control unit 342 controls the rotation of the rotating plate 332, for example, rotating the plate 332 at a rate of 1.0 s / cycle. The detector control unit 344 controls the X-ray detection performed by the X-ray detector 320, for example, detecting X-rays at a rate of 0.4 degrees / cycle. The stage control unit 343 controls the movement of the stage 102.

[0032] The overall control unit 400 includes a CPU (Central Processing Unit) 401, a memory 402, a storage device 403, a control imaging control unit 340, and performs various processes on projection data acquired by the X-ray detector 320. For example, the overall control unit 400 performs processing to reconstruct a tomographic image from projection data acquired according to imaging conditions set by the input device 210. The tomographic image can be reconstructed per energy cell. Furthermore, the reconstructed tomographic image and the projection data used in the reconstruction process can be displayed on the monitor 220, stored in the storage device 403, or processed per energy cell. Moreover, by imaging a combination of multiple substrate materials with known composition and thickness, material identification data used in the creation of a material identification image is obtained.

[0033] use Figure 3 The material identification data 520 used in the creation of the material identification image is explained below. The material identification data 520 is a variety of X-ray energy spectra obtained using a correction member 510 composed of multiple substrate materials with known composition and thickness. The correction member 510 is, for example, a member composed of a first substrate material 511 and a second substrate material 512 with a uniform thickness and an arc shape centered on the X-ray focus 501. The first substrate material 511 uses a material with a relatively small effective atomic number, such as acrylic resin or polyethylene, while the second substrate material 512 uses a material with a relatively large effective atomic number, such as aluminum, calcium mixtures, iodine mixtures, or tin. Furthermore, the first substrate material 511 and the second substrate material 512 each have multiple different thicknesses. For example, if the thickness of the first substrate material 511 is type J and the thickness of the second substrate material 512 is type K, then a correction member 510 of type J×K is used, and for each combination, an X-ray energy spectrum is obtained for each detection element 321. Figure 3 Since J=3 and K=3, nine types of X-ray energy spectra are shown as material identification data 520. Furthermore, the material identification data 520 is obtained in advance before the photograph of the subject 101.

[0034] The obtained substance identification data 520 can, for example, be used as... Figure 4 The substance identification map 521 shown is stored in the storage device 403. Figure 4 The exemplified matter discrimination map 521 is the result of detecting X-rays transmitted through a correction member 510 composed of acrylic resin with a thickness of 0 mm to 200 mm and aluminum with a thickness of 0 mm to 8 mm by splitting them into three energy chambers. Furthermore, the matter discrimination map 521 is not limited to... Figure 4 The example table can also be represented by charts, mathematical expressions, etc.

[0035] The steps for creating a material identification image are explained using material identification data 520. Projection data of the subject 101 acquired by the imaging unit 300 at various projection angles have X-ray energy spectra for each detection element 321. Therefore, the X-ray energy spectrum from the material identification data 520 is searched for the energy spectrum whose shape is closest to the X-ray energy spectrum of each detection element 321, and a combination of the thickness of the substrate material corresponding to the searched X-ray energy spectrum is obtained. That is, if the first substrate material 511 is acrylic resin and the second substrate material 512 is aluminum, the thickness of the acrylic resin and the thickness of the aluminum are obtained for each detection element 321 at various projection angles, and the projection data of the acrylic resin and the aluminum are obtained. Then, a tomographic image of each substrate material is reconstructed from the obtained projection data of each substrate material to create a material identification image.

[0036] use Figure 5 This is used to illustrate the banded artifacts produced in material discrimination images. Figure 5 The image shown is an aluminum image 540 and an acrylic resin image 550, used as material discrimination images created using projection data of a phantom 530 simulating a subject containing regions with high X-ray absorption coefficients. The phantom 530 contains rod-shaped aluminum 531 within a cylindrical acrylic resin 532. The aluminum 531 simulates regions with high X-ray absorption coefficients. The acrylic resin 532 simulates regions other than those with high X-ray absorption coefficients.

[0037] In the aluminum image 540, which represents the distribution of aluminum, only the aluminum region 541 corresponding to the aluminum 531 in the phantom 530 should be displayed. However, between the aluminum regions 541 corresponding to areas with high X-ray absorption coefficients, a band-like artifact, namely a dark band 542, is generated with a pixel value lower than the surrounding area. Similarly, in the acrylic resin image 550, which represents the distribution of acrylic resin, the pixel values ​​of the acrylic resin region 551 corresponding to the acrylic resin 532 should be uniform. However, between the regions corresponding to the aluminum 531, a band-like artifact, namely a bright band 552, is generated with a pixel value higher than the surrounding area. Medical images containing dark bands 542 and bright bands 552 can lead to a decrease in diagnostic accuracy.

[0038] The dark band 542 and the bright band 552 are generated because the proportion of scattered rays in the X-rays incident on the detection element 321a located directly below the aluminum 531 is greater than the proportion of scattered rays in the material identification data 520. That is, in Figure 5 In the process, most of the scattered rays generated by the acrylic resin 532 are incident on the detection element 321a; in contrast, in Figure 3 In this process, the scattered rays generated by the acrylic resin, which serves as the first substrate material 511, are absorbed by the aluminum, which serves as the second substrate material 512, resulting in fewer scattered rays incident on the detection element 321a. Consequently, even if the thickness combination is the same in aluminum 531 and acrylic resin 532, and in the first substrate material 511 and second substrate material 512, differences will occur in the X-ray energy spectrum of the detection element 321a, leading to errors in the thickness obtained through material identification. More specifically, due to... Figure 5 X-ray energy spectrum and Figure 3 In contrast, the X-ray intensity on the low-energy side becomes slightly higher. Therefore, the aluminum is made thinner and the acrylic resin is made thicker than it actually is, resulting in dark band 542 and bright band 552 due to the thickness error. In addition, the difference between the average energy of the X-rays after passing through the phantom 530 and the average energy of the X-rays after passing through the correction member 510 is within the acceptable range.

[0039] While one of the dark band 542 and the bright band 552 can be corrected by beam hardening correction, the other artifact cannot be corrected. Therefore, in Example 1, the band artifact of the other material discrimination image is corrected using the correction amount used to correct the band artifact of one material discrimination image, namely the first correction amount, and the correction amount calculated based on the average energy of the X-rays passing through the subject, namely the second correction amount.

[0040] use Figure 6 The functional blocks of Embodiment 1 will be described below. These functional blocks can be constructed using dedicated hardware or software operating on the CPU 401. In the following description, the case where the functional blocks of Embodiment 1 are constructed using software will be explained. Embodiment 1 includes a first correction unit 601, an energy calculation unit 602, and a second correction unit 603. Each structural unit will be described below.

[0041] The first correction unit 601 corrects the banded artifacts in the first substance identification image and calculates the correction amount of the banded artifacts in the first substance identification image, i.e., the first correction amount. The correction of the banded artifacts in the first substance identification image can be any method, such as the method described in Patent Document 1. The first correction amount is obtained by subtracting the substance identification image before correction from the substance identification image after correction of the banded artifacts.

[0042] The energy calculation unit 602 calculates the average energy of the X-rays transmitted through the subject 101. The average energy can be calculated based on a virtual monochromatic image created using a material discrimination image containing band artifacts, which serves as an image at a specific X-ray energy. That is, the virtual monochromatic image with the smallest band artifacts is extracted from multiple virtual monochromatic images, and the X-ray energy corresponding to the extracted virtual monochromatic image is calculated as the average energy. Alternatively, the average energy can be calculated from the X-ray energy spectrum of the projection data of the subject 101.

[0043] The second correction unit 603 calculates a second correction based on the first correction and the average energy, and uses the second correction to correct banding artifacts in the image of another substance. The second correction, COR2, is calculated, for example, using the following formula.

[0044] COR2=μ1(E)·ρ1·COR1 / (μ2(E)·ρ2)…(Formula 1)

[0045] Here, μ1(E) is the X-ray attenuation coefficient of the first substance at X-ray energy E, ρ1 is the density of the first substance, COR1 is the first correction, μ2(E) is the X-ray attenuation coefficient of the second substance at X-ray energy E, ρ2 is the density of the second substance, and E uses the average energy.

[0046] use Figure 7 Here is an example of the processing flow of Example 1, explained step by step.

[0047] (S701)

[0048] The overall control unit 400 controls the shooting control unit 340 to obtain the projection data of the subject 101.

[0049] (S702)

[0050] The overall control unit 400 performs substance identification based on the projection data acquired in S701, and acquires projection data of the first substance and the second substance. For example, the substance identification uses pre-acquired substance identification data 520.

[0051] (S703)

[0052] The overall control unit 400 reconstructs the projection data of the first substance and the projection data of the second substance obtained in S702 to create a first substance identification image and a second substance identification image.

[0053] (S704)

[0054] The overall control unit 400 uses the first and second substance identification images obtained in S703 to create multiple virtual monochrome images. The pixel value V of the virtual monochrome image at X-ray energy E is calculated, for example, using the following formula.

[0055] V=μ1(E)·ρ1·M1+μ2(E)·ρ2·M2…(Formula 2)

[0056] Here, μ1(E) is the X-ray attenuation coefficient of the first substance at X-ray energy E, ρ1 is the density of the first substance, M1 is the pixel value of the first substance's identification image, μ2(E) is the X-ray attenuation coefficient of the second substance at X-ray energy E, ρ2 is the density of the second substance, and M1 is the pixel value of the second substance's identification image.

[0057] (S705)

[0058] The overall control unit 400 determines whether the first substance identification image, the second substance identification image, or the virtual monochrome image created in S704 contains banding artifacts. If banding artifacts are present, the process proceeds to S706; otherwise, the process proceeds to S708. The presence or absence of banding artifacts can be determined by a pre-created arbiter using machine learning, or by the operator.

[0059] (S706)

[0060] The overall control unit 400 corrects banded artifacts in the material identification image.

[0061] use Figure 8 This is an example of the process for correcting the material identification image in S706, step by step.

[0062] (S801)

[0063] The first correction unit 601 corrects banding artifacts in a material discrimination image, such as the dark band 542 in an aluminum image 540. Banding artifacts are corrected, for example, by beam hardening correction.

[0064] (S802)

[0065] The first correction unit 601 calculates the correction amount for the band artifacts in a material discrimination image, i.e., the first correction amount. The first correction amount is calculated by subtracting the uncorrected image, i.e., the aluminum image 540, from the corrected image obtained in S801.

[0066] (S803)

[0067] The energy calculation unit 602 calculates the average energy of the X-rays transmitted through the subject 101. For example, the virtual monochrome image with the smallest band artifact is extracted from the multiple virtual monochrome images created in S704, and the X-ray energy corresponding to the extracted virtual monochrome image is calculated as the average energy. Alternatively, the average energy can be calculated from the X-ray energy spectrum of the projection data obtained in S701.

[0068] (S804)

[0069] The second correction unit 603 calculates the second correction amount based on the first correction amount calculated in S802 and the average energy calculated in S803. The second correction amount is calculated, for example, using (Equation 1).

[0070] (S805)

[0071] The second correction unit 603 uses the second correction amount calculated in S804 to correct band artifacts in another substance identification image, such as correcting the bright band 552 in the acrylic resin image 550. That is, by adding the second correction amount to another substance identification image, such as the acrylic resin image 550, the band artifact, i.e., the bright band 552, in the other substance identification image is corrected.

[0072] pass Figure 8 The illustrated correction process works by correcting banded artifacts in one material discrimination image for banded artifacts in another. Return to... Figure 7 Explanation.

[0073] (S707)

[0074] The overall control unit 400 uses the first and second substance identification images, which have been corrected in S706, to create a virtual monochrome image. The pixel values ​​of the virtual monochrome image are calculated, for example, using (Equation 2).

[0075] (S708)

[0076] The overall control unit 400 saves the first substance identification image, the second substance identification image, or the virtual monochrome image created in S705 and S707, which have been corrected in S706, to the storage device 403. The substance identification image and the virtual monochrome image stored in the storage device 403 are displayed on the monitor 220 as needed and used in diagnostics.

[0077] pass Figure 7 The illustrated processing flow generates material discrimination images and virtual monochrome images that have been corrected for band artifacts, thereby improving diagnostic accuracy.

[0078] The embodiments of the present invention have been described above. However, the present invention is not limited to the above embodiments, and the constituent elements can be modified and embodied without departing from the spirit of the invention. Furthermore, the multiple constituent elements disclosed in the above embodiments can be appropriately combined. Moreover, several constituent elements can be deleted from all the constituent elements shown in the above embodiments.

Claims

1. A PCCT device that acquires projection data divided into multiple energy chambers by irradiating a subject with X-rays, the PCCT device being characterized by comprising: The first correction unit corrects the band artifacts in the first material identification image among multiple material identification images made from the projection data, and calculates the correction amount of the band artifacts, which is the first correction amount. The energy calculation unit calculates the average energy of the X-rays that pass through the subject; The second correction unit uses a correction amount calculated based on the first correction amount and the average energy, i.e., the second correction amount, to correct the band artifacts in the second substance discrimination image.

2. The PCCT device according to claim 1, characterized in that, The energy calculation unit calculates the average energy based on multiple virtual monochrome images created using the first substance identification image and the second substance identification image.

3. The PCCT device according to claim 2, characterized in that, The energy calculation unit extracts the virtual monochrome image with the smallest band artifact from the plurality of virtual monochrome images, and uses the X-ray energy corresponding to the extracted virtual monochrome image as the average energy.

4. The PCCT device according to claim 1, characterized in that, The energy calculation unit calculates the average energy from the X-ray energy spectrum of the projection data.

5. The PCCT device according to claim 1, characterized in that, When the X-ray attenuation coefficient of the first substance at average energy E is μ1(E), the density of the first substance is ρ1, the X-ray attenuation coefficient of the second substance at average energy E is μ2(E), the density of the second substance is ρ2, and the first correction is COR1, the second correction COR2 is calculated by COR2=μ1(E)·ρ1·COR1 / (μ2(E)·ρ2).

6. The PCCT device according to claim 1, characterized in that, The first correction unit corrects the band artifacts in the first substance identification image by beam hardening correction.

Images