CT image processing device, program, and CT image processing method

Abstract

The present invention comprises: a medical image acquisition unit 21, a reconstructed image generation unit 22, and a high-absorber region extraction unit 23 that specify the reconstructed image coordinates of an artifact source on a CT three-dimensional reconstructed image based on a CT projection image; a coordinate conversion unit 24 that directly converts the reconstructed image coordinates into CT projection image coordinates on the CT projection image; a high-absorber region generation unit 25 that generates the region of the artifact source on the CT projection image on the basis of the converted CT projection image coordinates; and a high-absorber region interpolation unit 26 that interpolates the pixel values of the region generated by the high-absorber region generation unit 25, with the pixel values of the other regions. Thus, a CT image processing device, a program, and a CT image processing method are provided such that a high artifact reduction effect and the prevention of occurrence of secondary artifacts are both achieved at low calculation cost in processing on a CT image.

Description

CT Image Processing Apparatus, Program, and CT Image Processing Method

[0001] The present invention relates to a CT image processing apparatus, a program, and a CT image processing method for processing CT image information.

[0002] In Patent Document 1, as an example of an X-ray CT apparatus that efficiently identifies abnormal detection elements, it includes a collection unit, a first identification unit, a second identification unit, and a third identification unit. The collection unit includes an X-ray detector including a plurality of detection elements, and collects a projection data set for each projection angle by detecting X-rays transmitted through the subject with the X-ray detector. The first identification unit identifies a first region corresponding to an artifact derived from an abnormal element included in the X-ray detector in an image reconstructed based on the projection data set. The second identification unit identifies a second region that contributed to the imaging of the first region in the projection data sets corresponding to each of the plurality of projection angles. The third identification unit identifies an abnormal element based on the second regions identified for each of the plurality of projection angles. This is described.

[0003] Patent No. 7206163

[0004] It is a technology related to reducing the high absorber artifact component generated due to a high absorber (a high radiation absorber such as metal) in a subject in an image obtained by an X-ray CT (Computed Tomography) apparatus for image diagnosis or an image obtained by an X-ray CT apparatus during radiation therapy.

[0005] In a radiation therapy system, a technology has been developed that mounts a cone beam CT apparatus (CBCT (Cone Beam Computed Tomography) apparatus) to acquire a CT image and uses it for correcting the subject position and monitoring the treatment target so as to match a pre-established radiation therapy plan.

[0006] A CBCT apparatus is an apparatus that calculates the X-ray absorption coefficient in a subject from projection data obtained by photographing the subject from multiple directions and obtains a CT image.

[0007] In CT scanners, the X-ray absorption coefficient is generally replaced with the Hounsfield Unit value, which is normalized for air and water (air = -1000, water = 0).

[0008] Here, we will explain high-absorption artifacts caused by high-absorption materials. When imaging with an X-ray CT scanner, if there are high-absorption materials with a large X-ray absorption coefficient, such as metals or bones, in the subject's body, the X-rays are rapidly absorbed, mainly in the low-energy band, as they pass through these materials. Therefore, if a high-absorption material is located in the straight path between the X-ray source and the detector, the obtained measurement projection data may be unreliable. Linear artifacts (hereinafter referred to as "high-absorption artifacts") occur in CT images reconstructed based on unreliable measurement projection data, and these are referred to here as high-absorption artifacts. These high-absorption artifacts are a major factor in significantly reducing the accuracy of clinical diagnostics.

[0009] During radiation therapy, the radiation dose is determined based on planning CT images taken before the start of treatment. The planning CT images are then aligned with CT images taken immediately before treatment, and the patient's position on the treatment table is set so that the affected area of ​​the patient aligns with the radiation irradiation area. However, the high-attenuation artifacts mentioned above can reduce the accuracy of radiation dose determination and patient positioning during radiation therapy.

[0010] Furthermore, the CT images used for treatment planning are often from a fan-beam CT scanner, similar to those used for diagnostic purposes. This often results in differences in the location and intensity of high-attenuation artifacts compared to the CBCT used for CT imaging immediately before treatment. Consequently, the accuracy of alignment decreases, leading to discrepancies between the treatment target and the irradiation position.

[0011] Furthermore, if the treatment target is located in a position that overlaps with artifacts on the CBCT image, it may become difficult to pinpoint the location of the treatment target, potentially leading to a decrease in the accuracy of irradiation.

[0012] As a means of reducing artifacts using high-absorbent materials, there are methods described in Patent Document 1, etc. However, even with the methods described in Patent Document 1, etc., it is difficult to achieve sufficient reduction effect and prevent the occurrence of secondary artifacts while reducing computational costs, and new methods are awaited.

[0013] The present invention provides a CT image processing apparatus, a program, and a CT image processing method that achieve both high artifact reduction and prevention of secondary artifacts with low computational cost in processing CT images.

[0014] The present invention includes multiple means for solving the above problems, but one example is a CT image processing apparatus for processing CT images of a specimen, comprising: a coordinate identification unit for identifying the reconstructed image coordinates of an artifact source on a CT 3D reconstructed image based on a CT projection image; a coordinate transformation unit for directly converting the reconstructed image coordinates to CT projection image coordinates on the CT projection image; a region generation unit for generating the region of the artifact source on the CT projection image based on the transformed CT projection image coordinates; and an interpolation unit for interpolating the pixel values ​​of the region generated by the region generation unit with the pixel values ​​of other regions.

[0015] According to the present invention, in processing CT images, it is possible to achieve both a high artifact reduction effect and prevention of secondary artifact generation with low computational cost. Other issues, configurations, and effects will be clarified by the following description of the embodiments.

[0016] This figure shows an overview of the radiotherapy system of the embodiment. This figure shows an example of the system configuration including the medical image processing device according to the embodiment. This is a flowchart explaining an example of the processing overview according to the embodiment. This figure shows the relationship between the reconstructed image coordinate system and the projection image coordinate system according to the embodiment. This figure shows an example of the configuration of the reconstructed image generation unit according to the embodiment. This figure explains the flowchart when outputting a high-attenuation region reconstructed image according to the embodiment.

[0017] Embodiments of the CT image processing apparatus, program, and CT image processing method of the present invention will be described with reference to Figures 1 to 6. In the drawings used herein, the same or corresponding components are denoted by the same or similar reference numerals, and repeated descriptions of these components may be omitted.

[0018] First, the overall configuration of a radiotherapy system to which a CT image processing device is suitably applied will be described using Figure 1. Figure 1 is a diagram showing the schematic configuration of a radiotherapy system according to this embodiment.

[0019] The radiotherapy system 150 shown in Figure 1 is mainly composed of a bed 2 that supports the patient (subject 1), a rotating support device 3 surrounding the bed 2, an X-ray tube 5, an FPD (Flat Panel Display) 4, a medical image processing device 11, a radiation control device 120, a therapeutic radiation generator 130, and a therapeutic radiation irradiation device 140.

[0020] The therapeutic radiation generator 130 is composed of equipment that generates various types of radiation, such as X-rays, proton beams, and heavy ion beams such as deuterium and carbon. Its configuration varies depending on the type of radiation to be generated, and various known configurations can be adopted.

[0021] In configurations that irradiate with proton beams or heavy ion beams, the therapeutic radiation generator 130 includes, for example, an ion source, a linac, and a synchrotron. The synchrotron includes a deflection magnet, a quadrupole magnet, a radio frequency accelerator, a radio frequency emitter, and an emitter deflector.

[0022] The ion source is connected to a linac, which is connected to a synchrotron. Particles generated from the ion source are pre-accelerated by the linac and then injected into the synchrotron. The particle beam, further accelerated in the synchrotron, is then emitted into the transport system.

[0023] The transport system includes multiple deflection electromagnets and a quadrupole electromagnet (not shown) and connects the synchrotron to the therapeutic radiation irradiation device 140. Furthermore, a portion of the therapeutic radiation generator 130 (the transport system) and the therapeutic radiation irradiation device 140 are mounted on a cylindrical rotating support device 3 and can rotate together with the rotating support device 3. The particle beam emitted from the synchrotron passes through the therapeutic radiation generator 130, is focused by the quadrupole electromagnet, and then its direction is changed by the deflection electromagnet before it enters the therapeutic radiation irradiation device 140.

[0024] The therapeutic radiation irradiation device 140 is equipped with two scanning electromagnets, a dose monitor, and a position monitor. The scanning electromagnets are positioned perpendicular to each other and generate a magnetic field by excitation current, which can deflect the particle beam so that it reaches a desired position in a plane perpendicular to the beam axis at the target location. The dose monitor measures the amount of irradiated particle beam. The position monitor can detect the position through which the particle beam has passed. The particle beam that has passed through the therapeutic radiation irradiation device 140 reaches the target in the patient 1. When treating patients with cancer or the like, the patient 1 is the patient, and the target is the tumor or the like.

[0025] The platform on which the subject 1 is placed is called the bed 2. Based on instructions from the radiation control device 120, the bed 2 can move in the direction of three orthogonal axes and can also rotate around each axis. Through these movements and rotations, the position of the subject 1, including the target, can be moved to the desired position.

[0026] The radiation control device 120 controls the irradiation and stopping of particle beams, as well as the irradiation and stopping of X-rays for acquiring CT images, and controls the X-ray imaging device and the rotation support device 3. It is electrically connected to the X-ray imaging device, the therapeutic radiation generator 130, the therapeutic radiation irradiation device 140, the bed 2, the input unit 10, etc., in order to control equipment such as the X-ray imaging device, the therapeutic radiation generator 130, and the therapeutic radiation irradiation device 140.

[0027] The CT scanner comprises a bed 2 for supporting a subject 1, a rotating support device 3 surrounding the bed 2, one or more X-ray imaging devices attached to the rotating support device 3 so as to rotate integrally with the rotating support device 3, a rotating device (not shown) for rotating the rotating support device 3, and a medical image processing device 11.

[0028] The X-ray imaging apparatus consists of one or more pairs of X-ray tubes 5 and FPDs 4, and an X-ray diaphragm 6.

[0029] The X-ray tube 5 generates X-rays by receiving a high voltage and filament current from a high voltage generation unit (not shown). The time interval for irradiating the subject 1 with X-rays is, for example, 10 times per second. The high voltage generation unit applies a high voltage to the X-ray tube 5 and supplies tube current according to the imaging parameters determined by the medical image processing device 11.

[0030] An X-ray diaphragm 6 is attached to the X-ray irradiation port side of the X-ray tube 5. The X-ray diaphragm 6 limits the irradiation field of the X-rays generated from the X-ray tube 5. Specifically, the X-ray diaphragm 6 movably supports multiple diaphragm blades made of a material that shields X-rays (such as lead). By adjusting the position of the multiple diaphragm blades, the size and shape of the X-ray irradiation field are changed. The X-ray diaphragm 6 moves its diaphragm blades in response to a drive signal supplied from the control unit.

[0031] The FPD 4 detects X-rays generated from the X-ray tube 5 and transmitted through the subject 1, and generates a current signal corresponding to the intensity of the detected X-rays. A data acquisition circuit (not shown) is connected to the FPD 4 and collects the current signal output from the FPD 4. The data acquisition circuit amplifies the collected current signal and converts the amplified current signal into digital data to generate X-ray projection data, which is a digital signal. The X-ray projection data is transferred to the medical image processing device 11 via the input unit 10 and stored in the medical information storage unit 12 along with time-series information.

[0032] As the rotating support device 3 rotates, it irradiates X-rays from the X-ray tube 5 and generates X-ray projection data with the FPD 4, thereby acquiring the X-ray projection data necessary for reconstructing the cone-beam CT image.

[0033] The medical image processing device 11 is a device that can perform high-absorber artifact reduction processing on CT images with low computational cost, achieving both high reduction effect and prevention of secondary artifact generation. Figure 2 shows its details. Figure 2 is a diagram showing an example of a system configuration including the medical image processing device according to this embodiment.

[0034] The medical image processing device 11 shown in Figure 2 is a device that processes CT images of a subject 1 and reconstructs CT images from acquired X-ray projection data. It also sets the imaging parameters of the X-ray tube 5 and FPD 4. As shown in Figure 2, the medical image processing device 11 is electrically connected to an input unit 10, a medical information storage unit 12, and an output unit 13.

[0035] The input unit 10 receives input from the operator or a higher-level system and transmits a signal to the medical image processing device 11. The medical information storage unit 12 stores various information necessary for or used in medical images and reconstruction, and transmits and receives it with the medical image processing device 11. The output unit 13 is the part that outputs using the medical images obtained from the medical image processing device 11, and displays the generated X-ray projection data and the created CT image on the screen.

[0036] As shown in Figure 2, this medical image processing apparatus 11 includes a medical image acquisition unit 21, a reconstructed image generation unit 22, a high-absorbent region extraction unit 23, a coordinate transformation unit 24, a high-absorbent region generation unit 25, and a high-absorbent region interpolation unit 26.

[0037] The medical image acquisition unit 21 acquires a projected image from the medical information storage unit 12. The reconstructed image generation unit 22 generates a reconstructed image from the projected image. The high-absorbent region extraction unit 23 extracts high-absorbent regions from the reconstructed image. Together, the medical image acquisition unit 21, the reconstructed image generation unit 22, and the high-absorbent region extraction unit 23 constitute a coordinate identification unit that identifies the reconstructed image coordinates of artifact sources on the CT 3D reconstructed image based on the CT projected image, and preferably acts as the main execution unit for the coordinate identification procedure and coordinate identification step.

[0038] The coordinate transformation unit 24 is the part that directly transforms the reconstructed image coordinates into CT projection image coordinates on the CT projection image, and transforms the extracted high-absorbent regions on the reconstruction into high-absorbent regions on the projection image. Preferably, this coordinate transformation unit 24 is the main body that executes the coordinate transformation procedure and coordinate transformation steps.

[0039] The high-absorbent region generation unit 25 is the part that generates artifact source regions on the CT projection image based on the converted CT projection image coordinates, and identifies high-absorbent regions on the projection image. Preferably, this high-absorbent region generation unit 25 is the main body that executes the region generation procedure and region generation steps.

[0040] The high-absorbent region interpolation unit 26 is the part that interpolates the pixel values ​​of the region generated by the high-absorbent region generation unit 25 with the pixel values ​​of the other regions, thereby interpolating the high-absorbent region on the projected image with the information of pixels in the other regions. Various known interpolation methods can be used. Preferably, this high-absorbent region interpolation unit 26 is the main body that executes the interpolation procedure and interpolation steps.

[0041] In this invention, the "artifact source" is a high-radiation absorber or high-luminosity material, for example, a region where the CT value is greater than a predetermined threshold in absolute value, more specifically, a region where the CT value is approximately 1000 or more in absolute value.

[0042] The medical image processing device 11 and the radiation control device 120 described above may each have a central processing unit (CPU) and memory connected to this CPU, or they may be configured as a single computer, and are not particularly limited in any way.

[0043] Furthermore, the control processes for the actions to be performed may be combined into a single program, divided into multiple programs, or a combination of these.

[0044] Some or all of the programs contained within each device may be implemented using dedicated hardware, or they may be modularized. Furthermore, various programs may be installed on each device via a program distribution server or external storage media, or existing devices may be updated.

[0045] Further, each device may be connected by a wired or wireless network as an independent device, or two or more of them may be integrated.

[0046] Next, an example of the processing by the medical image processing apparatus 11 having the system configuration shown in FIG. 2 will be described with reference to FIG. 3. FIG. 3 is a flowchart for explaining an example of the outline of the processing according to the embodiment.

[0047] As shown in FIG. 3, first, the medical image acquisition unit 21 of the medical image processing apparatus 11 acquires a medical image from the medical information storage unit 12 according to an instruction from the system or the operator (step S101).

[0048] Next, the reconstructed image generation unit 22 generates a three-dimensional reconstructed image from the acquired projection image (step S102).

[0049] Next, the high absorber region extraction unit 23 specifies the coordinates (positions) of the high absorber region on the three-dimensional reconstructed image (step S103).

[0050] These steps S101 to S103 correspond to a coordinate specifying step for specifying the reconstructed image coordinates of the artifact source on the CT three-dimensional reconstructed image based on the CT projection image.

[0051] Next, the coordinate conversion unit 24 converts the high absorber region in the reconstructed image coordinate system into the projection image coordinate system (step S104). This step S104 corresponds to a coordinate conversion step for directly converting the reconstructed image coordinates into the CT projection image coordinates on the CT projection image.

[0052] The processing performed by the coordinate conversion unit 24 in this step S104 will be described with reference to FIG. 4. FIG. 4 is a diagram showing the relationship between the reconstructed image coordinate system and the projection image coordinate system according to the embodiment.

[0053] In the conventional method, the metal region extracted by the high absorber region extraction unit 23 is orthographically projected to specify the metal region on the projection image. In contrast, in the processing of the present invention, this is realized by coordinate conversion without orthographically projecting this metal region.

[0054] Figure 4 shows the relationship between the X-ray source 45, the subject 1 and FPD 4, and the volume coordinate system 41, which is the coordinate system for the subject and reconstruction, and the FPD coordinate system 44, which is the coordinate system for the projected image.

[0055] In the volume coordinate system, the information of which pixel in the projected image (xr, yr) is reflected can be obtained by transforming it into (xp, yp) as shown in equation (1).

[0056]

[0057] In equation (1), SID (Source to Image Distance) represents the distance from the X-ray source 45 to the subject 1, and SOD (Source to Object Distance) represents the distance from the X-ray source 45 to the FPD 4.

[0058] Here, it is common for the number of pixels, or resolution, for the same area to differ significantly between the projected image and the reconstructed image. Although it varies depending on the imaging system and reconstruction conditions, the projection image often has around 1000 to 2000 pixels per side, while the reconstructed image has around 512 pixels.

[0059] Thus, it can be seen that the projected image has a resolution of about two to four times that of the reconstructed image. For this reason, even if the coordinate transformation unit 24 calculates the pixel positions on the projected image from the pixel positions extracted from the reconstructed image, only information with a resolution of about one-quarter to one-half can be obtained. To bridge this gap, the high-absorbent region generation unit 25 generates high-absorbent regions on the projected image.

[0060] The point cloud on the projected image obtained by the coordinate transformation unit 24 often consists of discrete points, meaning that the high-absorbent region, which was a set of adjacent pixels in the reconstruction, often appears as multiple non-adjacent points on the projected image. It is thought that by generating a set of adjacent points on the projected image, the high-absorbent region on the projected image can be identified.

[0061] One possible method is to pre-calculate the difference in resolution and unfold it onto the projected image as a square or circle centered on the coordinate-transformed point. Alternatively, the information of adjacent pixels on the reconstructed image can be attached to each point before transforming it onto the projected image, and after transformation, the adjacent relationship on the reconstructed image can be referenced to make linear projection image pixels connecting adjacent points into high-absorbent regions.

[0062] Thus, the high-absorbent region generation unit 25 can generate regions from the coordinate group transformed by the coordinate transformation unit 24 by one or more of the following: geometric shape deformation, deformation based on pixel information on the CT projection image.

[0063] Regardless of the method used, adjacent point clouds will be generated as high-absorbent regions on the projected image, but the contour shape of the high-absorbent region as a collection of point clouds may be distorted, linear, or sparse. The high-absorbent region on the projected image at this point is designated as a preliminary high-absorbent region.

[0064] For areas with high absorption, morphological processing can be performed to create geometrically smooth and dense regions. Morphological processing in this context refers to, for example, closing processing. Closing processing involves performing several dilation processes to add pixels from the surrounding region to the target region, and then performing the same number of stenosis processes to remove pixels that form the contour of the dilated region, thereby filling in the sparse areas within the region. This process of repeatedly performing dilation and stenosis to refine the shape of a region or remove noise within the region is called morphological processing.

[0065] Furthermore, it is possible to create a more precise region shape using information from the pixel values ​​on the projected image. In other words, by taking advantage of the characteristic that high absorbers appear as low brightness in the projected image, it is possible to designate only the low brightness regions of the preliminary high absorber region as high absorber regions, or to add only the low brightness pixels around the preliminary high absorber region as high absorber regions. In combination with the aforementioned closing process, it is also possible to perform the closing process only on low brightness pixels.

[0066] In this way, it is possible to obtain a smooth and dense high-absorbent region even on the projected image.

[0067] Returning to Figure 3, the high-absorbent region generation unit 25 then generates a high-absorbent region from the coordinate-transformed point cloud on the projection image (step S105). This step S105 corresponds to the region generation step, which generates the region of artifact source on the CT projection image based on the transformed CT projection image coordinates.

[0068] Next, the high-absorbent region interpolation unit 26 interpolates the high-absorbent region on the projected image with pixels from regions other than the high-absorbent region (step S106), and the output unit 13 outputs the interpolated projected image (step S107).

[0069] Of these, step S106 corresponds to an interpolation step in which the pixel values ​​of the region generated in the high-absorbent region generation unit 25 are interpolated with the pixel values ​​of the other regions.

[0070] Next, variations of the medical image processing device 11 will be explained using Figure 5. Figure 5 is a diagram showing an example of the configuration of the reconstructed image generation unit according to the embodiment.

[0071] As shown in Figure 5, the reconstructed image generation unit 22A of the medical image processing apparatus 11 includes a high-absorbent region identification unit 31 that identifies reconstructed image coordinates from a CT projection image, and a specific region reconstructed image generation unit 32 that generates and reconstructs an extracted projection image smaller in size than the CT projection image containing artifact sources.

[0072] The high-absorbent region identification unit 31 roughly identifies high-absorbent regions on the projected image. Even in a projected image, high-absorbent regions appear as lower-luminance images compared to other regions, making it possible to identify high-absorbent regions to some extent. Here, first, a rectangular region containing a low-luminance area considered to be a high-absorbent region is identified for the projected image at each angle.

[0073] The specific region reconstruction image generation unit 32 performs reconstruction processing using only rectangular regions of a common size in the projection images at each angle that encompass all the regions identified by the high-absorbent region identification unit 31.

[0074] In the case of a projection image, the image reflects the state in which the X-rays are attenuated by all the material in the straight line through which they pass. Therefore, even if there is no high absorber in that line, the brightness may be low due to reasons such as a long volume of material being passed through (distance traveled). Compared to the reconstructed image generated as a cross-sectional image, it is considered difficult to identify the detailed shape of the high absorber region. However, this is not a problem for this high absorber region identification unit 31, as its purpose is not to identify the detailed shape of the high absorber.

[0075] This reduces the computational cost compared to performing reconstruction using the entire projection image, resulting in lower computational costs. The degree of this effect depends on the size of the high-absorbent region within the entire projection image, i.e., the FOV (Field of View); the smaller the high-absorbent region, the greater the effect.

[0076] Next, the process for outputting a high-attenuation region reconstructed image using variations of the medical image processing device 11 shown in Figure 5 will be explained with reference to Figure 6. Figure 6 is a diagram illustrating the flow when outputting a high-attenuation region reconstructed image according to an embodiment.

[0077] Here, the high-absorbent region interpolation unit 26 may remove not only artifacts but also the region where the high absorbent material has been reconstructed. Therefore, depending on how the interpolated image is used, it may be necessary to restore the high-absorbent region on the reconstruction. For this reason, it is desirable for the specific region reconstruction image generation unit 32 to output the reconstructed image information of the reconstructed extracted projection image overlaid on the artifact source.

[0078] In this case, as shown in Figure 6, the specific region reconstruction image generation unit 32 reconstructs the region of the extracted projection image that includes the identified high-absorbent region, and stores the high-absorbent region on the reconstructed image along with its pixel value in memory or in the medical information storage unit 12 (step S103A).

[0079] Steps S101, S102, S104, S105, and S106 are the same as the steps in Figure 3.

[0080] Subsequently, as shown in Figure 6, in addition to the interpolated projection image, the output unit 13 outputs a reconstructed image of the high-attenuation region stored in memory or in the medical information storage unit 12 (step S107A). Furthermore, the high-attenuation region can also be restored in a subsequent process separate from this process.

[0081] In this way, by superimposing the output high-absorbent region reconstruction image onto the interpolated reconstruction image generated from the interpolated projection image, it becomes possible to restore the high-absorbent region on the reconstruction image.

[0082] Furthermore, the high-absorbent region reconstruction image output by the output unit 13 does not necessarily have to be the same as the high-absorbent region used in the high-absorbent region interpolation unit 26. In other words, the high-absorbent region reconstruction image extracted for this process and the high-absorbent region reconstruction image superimposed in a subsequent process may be different images.

[0083] Therefore, the high-absorbent region reconstruction image extracted for this process and another high-absorbent region reconstruction image are referred to as output high-absorbent region reconstruction images. The output high-absorbent region reconstruction images are extracted by the high-absorbent region extraction unit 23. Examples of such images include the high-absorbent region reconstruction image extracted for this process that has undergone smoothing or closing processing, or a high-absorbent region reconstruction image extracted based on a different threshold.

[0084] Here, as explained in the high-absorbent region generation unit 25, a high-absorbent region reconstruction image for output, shaped using a method that combines morphological processing and pixel information of the reconstructed image, is output from the output unit 13.

[0085] In this way, the specific region reconstruction image generation unit 32 can identify and output reconstruction image information for a second region that is different from the reconstruction image information for the first region identified for input to the coordinate transformation unit 24.

[0086] Next, the effects of this embodiment will be described.

[0087] The medical image processing apparatus 11, program, and CT image processing method for processing the CT image of the subject 1 in the above-described embodiment include a medical image acquisition unit 21 that identifies the reconstructed image coordinates of artifact sources on a CT 3D reconstructed image based on a CT projection image, reconstructed image generation units 22, 22A, a high-absorbent region extraction unit 23 (coordinate identification procedure, coordinate identification step), a coordinate transformation unit 24 (coordinate transformation procedure, coordinate transformation step) that directly converts the reconstructed image coordinates to CT projection image coordinates on the CT projection image, a high-absorbent region generation unit 25 (region generation procedure, region generation step) that generates regions of artifact sources on the CT projection image based on the transformed CT projection image coordinates, and a high-absorbent region interpolation unit 26 (interpolation procedure, interpolation step) that interpolates the pixel values ​​of the regions generated in the high-absorbent region generation unit 25 with the pixel values ​​of the other regions.

[0088] This eliminates the need for computationally expensive forward projection, allowing for a high level of artifact reduction even with lower computational costs, and preventing the occurrence of secondary artifacts, thus achieving a high level of absorber artifact reduction.

[0089] Furthermore, the reconstructed image generation unit 22A includes a high-absorbent region identification unit 31 that identifies the reconstructed image coordinates from the CT projection image, and a specific region reconstructed image generation unit 32 that generates and reconstructs an extracted projection image smaller in size than the CT projection image containing the artifact source. Therefore, the computational load can be further reduced compared to when the reconstruction process is performed using the entire projection image, thus enabling processing at a lower computational cost.

[0090] Furthermore, the high-absorbent region generation unit 25 can improve the accuracy of region generation by generating regions from the coordinate group transformed by the coordinate transformation unit 24 using one or more of the following methods: geometric shape deformation, deformation based on pixel information on the CT projection image, etc.

[0091] Furthermore, the specific region reconstruction image generation unit 32 outputs the reconstructed image information of the reconstructed extracted projection image overlaid on the artifact source, thereby enabling the restoration of the artifact source region as needed.

[0092] Furthermore, the specific region reconstruction image generation unit 32 can further speed up processing by identifying and outputting reconstruction image information for a second region that is different from the reconstruction image information for the first region identified for input to the coordinate transformation unit 24.

[0093] <Other Notes> The present invention is not limited to the embodiments described above, and various modifications are included. For example, the embodiments described above are explained in detail to make the present invention easier to understand, and are not necessarily limited to those having all the configurations described.

[0094] Furthermore, each of the above configurations, functions, processing units, processing means, etc., may be implemented in hardware, either partially or entirely, by designing them as integrated circuits, for example. Alternatively, each of the above configurations, functions, etc., may be implemented in software by having the processor interpret and execute programs that realize each function. Information such as programs, tables, and files that realize each function can be stored in memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.

[0095] Furthermore, the control lines and information lines shown are those deemed necessary for explanatory purposes, and not all control lines and information lines are necessarily shown in the actual product. In reality, it is safe to assume that almost all components are interconnected.

[0096] Furthermore, in another aspect of the present invention, in order to provide a CT image processing apparatus, program, and CT image processing method that achieve both a high artifact reduction effect and prevention of secondary artifact generation with low computational cost in processing CT images, the following aspects are provided.

[0097] (1) A CT image processing apparatus for processing CT images of a specimen, comprising: a coordinate identification unit that identifies the reconstructed image coordinates of an artifact source on a CT 3D reconstructed image based on a CT projection image, generates and reconstructs an extracted projection image smaller in size than the CT projection image containing the artifact source; and an interpolation unit that interpolates the pixel values ​​of the region of the artifact source with the pixel values ​​of the other regions based on the information of the identified reconstructed image coordinates. With this configuration, the amount of computation can be reduced compared to when the reconstruction process is performed using the entire projection image, so that a high reduction effect can be obtained with low computation cost, and secondary artifacts can be prevented.

[0098] (2) The CT image processing apparatus described in (1) further comprises a coordinate transformation unit that directly transforms the reconstructed image coordinates into CT projection image coordinates on the CT projection image, and a region generation unit that generates the region of the artifact source on the CT projection image based on the transformed CT projection image coordinates, wherein the interpolation unit interpolates the pixel values ​​of the region generated by the region generation unit with the pixel values ​​of the other regions. With this configuration, forward projection, which has a high computational cost, is not required, and processing with a lower computational cost can be achieved.

[0099] (3) In the CT image processing apparatus described in (2), the region generation unit generates the region from the coordinate group transformed by the coordinate transformation unit by one or more of the following: geometric shape deformation, deformation based on pixel information on the CT projection image. With this configuration, the accuracy of region generation can be improved.

[0100] (4) In the CT image processing apparatus described in any of (1) to (3), the coordinate identification unit outputs the reconstructed image information of the reconstructed extracted projection image superimposed on the artifact source. With this configuration, it is also possible to restore the region of the artifact source as needed.

[0101] (5) In the CT image processing apparatus described in (2) or (3), the coordinate identification unit identifies and outputs reconstructed image information of a second region that is different from the reconstructed image information of the first region identified for input to the coordinate transformation unit. With this configuration, processing can be made even faster.

[0102] 1...Subject 2...Bed 3...Rotating support device 4...FPD 5...X-ray tube 10...Input unit 11...Medical image processing unit (CT image processing unit) 12...Medical information storage unit 13...Output unit 21...Medical image acquisition unit (coordinate identification unit) 22, 22A...Reconstructed image generation unit (coordinate identification unit) 23...High-absorbent region extraction unit (coordinate identification unit) 24...Coordinate transformation unit 25...High-absorbent region generation unit (region generation unit) 26...High-absorbent region interpolation unit (interpolation unit) 31...High-absorbent region identification unit (region identification unit) 32...Specific region reconstructed image generation unit (specific region reconstructed image generation unit) 41...Volume coordinate system 44...FPD coordinate system (projection image coordinate system) 45...X-ray source 120...Radiation control device 130...Therapeutic radiation generator 140...Therapeutic radiation irradiation device 150...Radiation therapy system

Claims

1. A CT image processing apparatus for processing CT images of a specimen, comprising: a coordinate identification unit for identifying the reconstructed image coordinates of an artifact source on a CT 3D reconstructed image based on a CT projection image; a coordinate transformation unit for directly converting the reconstructed image coordinates to CT projection image coordinates on the CT projection image; a region generation unit for generating the region of the artifact source on the CT projection image based on the transformed CT projection image coordinates; and an interpolation unit for interpolating the pixel values ​​of the region generated by the region generation unit with the pixel values ​​of other regions.

2. The CT image processing apparatus according to claim 1, wherein the coordinate identification unit comprises a region identification unit that identifies the reconstructed image coordinates from the CT projection image, and a region reconstruction image generation unit that generates and reconstructs an extracted projection image smaller in size than the CT projection image containing the artifact source.

3. A CT image processing apparatus according to claim 1 or 2, wherein the region generation unit generates the region from the coordinate group transformed by the coordinate transformation unit by one or more of the following: geometric shape deformation, deformation based on pixel information on the CT projection image.

4. The CT image processing apparatus according to claim 2, wherein the specific region reconstruction image generation unit outputs the reconstructed image information of the reconstructed extracted projection image superimposed on the artifact source.

5. The CT image processing apparatus according to claim 4, wherein the specific region reconstruction image generation unit identifies and outputs reconstruction image information of a second region that is different from the reconstruction image information of a first region identified for input to the coordinate transformation unit.

6. A program executed in a CT image processing device that processes CT images of a specimen, the program causing the CT image processing device to execute: a coordinate identification procedure for identifying the reconstructed image coordinates of an artifact source on a CT 3D reconstructed image based on a CT projection image; a coordinate transformation procedure for directly converting the reconstructed image coordinates to CT projection image coordinates on the CT projection image; a region generation procedure for generating the region of the artifact source on the CT projection image based on the transformed CT projection image coordinates; and an interpolation procedure for interpolating the pixel values ​​of the region generated in the region generation procedure with the pixel values ​​of the other regions.

7. A CT image processing method for processing CT images of a specimen, comprising: a coordinate identification step for identifying the reconstructed image coordinates of an artifact source on a CT 3D reconstructed image based on a CT projection image; a coordinate transformation step for directly converting the reconstructed image coordinates to CT projection image coordinates on the CT projection image; a region generation step for generating a region of the artifact source on the CT projection image based on the transformed CT projection image coordinates; and an interpolation step for interpolating the pixel values ​​of the region generated in the region generation step with the pixel values ​​of other regions.

Images