IMAGE PROCESSING APPARATUS, IMAGE PROCESSING METHOD, AND IMAGE PROCESSING PROGRAM

Abstract

To provide an image processing device, an image processing method and an image processing program that are capable of accurately detecting a spicula in a short processing time.SOLUTION: An image processing device detects a spicula candidate region having a radial line structure from each of a plurality of tomographic images indicating a plurality of tomographic planes of an object, selects as a tomographic image group a plurality of the tomographic images corresponding to a plurality of the spicula candidate regions indicating the same spicula structure among the detected spicula candidate regions, and generates a composite two-dimensional image using the selected tomographic image group.SELECTED DRAWING: Figure 8

Description

[Technical field]

[0001] The present disclosure relates to an image processing device, an image processing method, and an image processing program. [Background technology]

[0002] In recent years, imaging diagnosis using a radiation imaging device (called mammography) for imaging the breast has been attracting attention in order to promote early detection of breast cancer. In addition, tomosynthesis imaging has been proposed in mammography, in which a radiation source is moved to irradiate the breast from multiple radiation source positions to perform imaging, and multiple projection images obtained by this are reconstructed to generate a tomographic image with a desired tomographic plane emphasized. In tomosynthesis imaging, multiple projection images are obtained by imaging the breast at multiple radiation source positions by moving the radiation source parallel to the radiation detector or moving it to draw a circular or elliptical arc according to the characteristics of the imaging device and the required tomographic image. Then, multiple projection images are generated by reconstructing the multiple projection images obtained using a backprojection method such as a simple backprojection method or a filtered backprojection method, or an iterative reconstruction method, etc.

[0003] By generating such tomographic images on multiple slices of the breast, it is possible to separate structures that overlap in the depth direction of the slices in the breast. This makes it possible to find abnormal areas such as lesions that are difficult to detect in 2D images obtained by conventional plain radiography, in which radiation is irradiated onto the subject from a predetermined direction (hereinafter referred to as "simple 2D images").

[0004] There is also known a technology for generating a pseudo two-dimensional image (hereinafter referred to as a "synthetic two-dimensional image") equivalent to a simple two-dimensional image by synthesizing multiple tomographic images obtained by tomosynthesis imaging and located at different distances (height direction positions) from the detection surface of a radiation detector toward the radiation source by using an addition method, an averaging method, a maximum value projection method, a minimum value projection method, or the like (see Patent Document 1).

[0005] In addition, in order to improve the identification of lesions, etc., a technology is known in which possible lesions, etc. are identified from 3D volume data, tagged in tomographic images, and a composite 2D image is generated by forward projecting multiple tagged tomographic images (see Patent Document 2). [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2014-128716 A [Patent Document 2] Special Publication No. 2015-515296 Summary of the Invention [Problem to be solved by the invention]

[0007] However, Patent Document 2 does not disclose the detection of spicules as candidates for lesions, etc. Spicules are radial linear structures that arise when breast cancer involves surrounding tissues during its development. Some normal breast glands also have radial structures that can be confused with spicules. Normally, it is not easy to distinguish between spicules and normal breast glands in a single tomographic image, so analysis is performed using three-dimensional images. Patent Document 2 also performs three-dimensional analysis to identify candidates for lesions, etc. from three-dimensional images (3D volume data). There is a problem with three-dimensional analysis in that it takes a long time to process.

[0008] The present disclosure has been made in consideration of the above circumstances, and aims to provide an image processing device, an image processing method, and an image processing program that enable spicules to be detected accurately in a short processing time. [Means for solving the problem]

[0009] The image processing device of the present disclosure is an image processing device having at least one processor, which detects spicule candidate regions having radial line structures from each of a plurality of tomographic images representing a plurality of tomographic planes of a subject, selects a plurality of tomographic images from the detected plurality of spicule candidate regions corresponding to a plurality of the spicule candidate regions representing the same spicule structure as a group of tomographic images, and generates a composite two-dimensional image using the selected group of tomographic images.

[0010] The processor preferably determines whether the spicule candidate regions are spicules by analyzing the composite two-dimensional image.

[0011] The processor preferably performs control to display the composite two-dimensional image, and also performs control to display, on the composite two-dimensional image, the result of the determination as to whether or not the spicule candidate region is a spicule.

[0012] It is preferable that the processor detects, in addition to spicule candidate regions, lesion candidate regions having structures other than radial line structures, and when a candidate region is not a spicule candidate region, selects a smaller number of tomographic images from the group of tomographic images to generate a composite two-dimensional image than when the candidate region is a spicule candidate region.

[0013] The image processing method disclosed herein is a process performed by a processor provided in an image processing device to detect spicule candidate regions having radial line structures from each of a plurality of tomographic images representing a plurality of tomographic planes of a subject, select a plurality of tomographic images from the detected plurality of spicule candidate regions corresponding to a plurality of the spicule candidate regions that represent the same spicule structure as a group of tomographic images, and generate a composite two-dimensional image using the selected group of tomographic images.

[0014] The image processing program disclosed herein causes a processor provided in an image processing device to execute a process of detecting spicule candidate regions having radial line structures from each of a plurality of tomographic images representing a plurality of tomographic planes of a subject, selecting a plurality of tomographic images from among the detected plurality of spicule candidate regions that correspond to a plurality of the spicule candidate regions that represent the same spicule structure as a group of tomographic images, and generating a composite two-dimensional image using the selected group of tomographic images. Effect of the Invention

[0015] According to the present disclosure, it is possible to provide an image processing device, an image processing method, and an image processing program that enable spicules to be detected accurately in a short processing time. [Brief description of the drawings]

[0016] [Figure 1] FIG. 1 is a schematic diagram illustrating the configuration of a radiation image capturing system. [Diagram 2] 2 is a view of the radiation image capturing device as seen in the direction of arrow A in FIG. 1. [Diagram 3] FIG. 2 is a block diagram showing an example of a hardware configuration of the image processing device. [Figure 4] FIG. 2 is a block diagram showing an example of a functional configuration of the image processing device. [Diagram 5] 11A and 11B are diagrams for explaining a process of acquiring a projection image. [Figure 6] FIG. 11 is a diagram for explaining a process for generating a tomographic image. [Figure 7] FIG. 2 is a diagram showing an example of a plurality of tomographic images. [Figure 8] FIG. 11 is a diagram for explaining a selection process of a group of tomographic images. [Figure 9] FIG. 11 is a diagram for explaining a process for generating a composite two-dimensional image. [Figure 10] 11A and 11B are diagrams for explaining a process of determining a spicule candidate region. [Figure 11] 11A and 11B are diagrams illustrating an example of a display state of a composite two-dimensional image and a determination result. [Figure 12]13 is a flowchart showing an example of a composite two-dimensional image generation process. [Figure 13] 11A and 11B are diagrams illustrating an example of a display state of a composite two-dimensional image and a determination result. [Figure 14] 11A and 11B are diagrams illustrating an example of a display state of a composite two-dimensional image and a determination result. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] Hereinafter, examples of embodiments for carrying out the technology of the present disclosure will be described in detail with reference to the drawings.

[0018] First, the configuration of a radiation image capturing system 100 according to this embodiment will be described with reference to Figures 1 and 2. As shown in Figures 1 and 2, the radiation image capturing system 100 captures a subject, that is, a breast M, from a plurality of radiation source positions to obtain a plurality of radiation images, i.e., a plurality of projection images, in order to perform tomosynthesis imaging of the breast and generate a tomographic image. The radiation image capturing system 100 includes a mammography device 1, a console 2, an image storage system 3, and an image processing device 4.

[0019] The mammography device 1 has an arm unit 12 connected to a base (not shown) by a rotation shaft 11. An imaging table 13 is attached to one end of the arm unit 12, and a radiation irradiation unit 14 is attached to the other end so as to face the imaging table 13. The arm unit 12 is configured so that only the end to which the radiation irradiation unit 14 is attached can rotate, which makes it possible to rotate only the radiation irradiation unit 14 while fixing the imaging table 13.

[0020] A radiation detector 15 such as a flat panel detector is provided inside the imaging stand 13. The radiation detector 15 has a radiation detection surface 15A. Also provided inside the imaging stand 13 are a circuit board and the like on which are provided a charge amplifier that converts a charge signal read out from the radiation detector 15 into a voltage signal, a correlated double sampling circuit that samples the voltage signal output from the charge amplifier, and an AD (Analog-to-Digital) conversion unit that converts an analog voltage signal into a digital signal.

[0021] A radiation source 16 is housed inside the radiation irradiation unit 14. The radiation source 16 emits X-rays as radiation. The timing at which the radiation source 16 irradiates radiation and the radiation generation conditions of the radiation source 16 (e.g., tube voltage, irradiation time, etc.) are controlled by the console 2.

[0022] The arm 12 is also provided with a compression plate 17 that is disposed above the imaging table 13 and presses down on the breast M, a support section 18 that supports the compression plate 17, and a movement mechanism 19 that moves the support section 18 in the vertical direction. The distance between the compression plate 17 and the imaging table 13, i.e., the compressed breast thickness, is input to the console 2.

[0023] The console 2 has a function of controlling the mammography apparatus 1 using imaging orders and various information acquired from a RIS (Radiology Information System) (not shown) or the like via a network such as a wireless communication LAN (Local Area Network) and using instructions given directly by a technician or the like. Specifically, the console 2 causes the mammography apparatus 1 to perform tomosynthesis imaging of the breast M, thereby acquiring a plurality of projection images as described below, and reconstructing the plurality of projection images to generate a plurality of tomographic images. As an example, in this embodiment, a server computer is used as the console 2.

[0024] The image storage system 3 is a system that stores image data such as radiographic images and tomographic images captured by the mammography device 1. The image storage system 3 extracts image data from the stored image data in response to requests from the console 2, image processing device 4, etc., and transmits the image data to the device that issued the request. A specific example of the image storage system 3 is a PACS (Picture Archiving and Communication System).

[0025] Next, a hardware configuration of the image processing device 4 according to this embodiment will be described with reference to Fig. 3. As shown in Fig. 3, the image processing device 4 includes a CPU (Central Processing Unit) 20, a memory 21 as a temporary storage area, and a non-volatile storage unit 22. The image processing device 4 also includes a display 23 such as a liquid crystal display, an input device 24 such as a keyboard and a mouse, and a network I / F (InterFace) 25 connected to a network. The CPU 20, the memory 21, the storage unit 22, the display 23, the input device 24, and the network I / F 25 are connected to a bus 27. The CPU 20 is an example of a "processor" according to the technology of the present disclosure.

[0026] The storage unit 22 is realized by a hard disk drive (HDD), a solid state drive (SSD), a flash memory, or the like. The storage unit 22 as a storage medium stores an image processing program 30. The CPU 20 reads the image processing program 30 from the storage unit 22, expands it in the memory 21, and executes the expanded image processing program 30.

[0027] Next, a functional configuration of the image processing device 4 according to this embodiment will be described with reference to Fig. 4. As shown in Fig. 4, the image processing device 4 includes an acquisition unit 40, a detection unit 42, a selection unit 44, a synthesis unit 46, a determination unit 48, and a display control unit 50. The CPU 20 executes the image processing program 30 to function as the acquisition unit 40, the detection unit 42, the selection unit 44, the synthesis unit 46, the determination unit 48, and the display control unit 50.

[0028] The acquisition unit 40 acquires a plurality of tomographic images generated by the console 2 causing the mammography apparatus 1 to perform tomosynthesis imaging. The acquisition unit 40 acquires a plurality of tomographic images from the console 2 or the image storage system 3 via the network I / F 25.

[0029] Here, the tomosynthesis imaging and generation of tomographic images in the console 2 will be described. When performing tomosynthesis imaging for generating tomographic images, the console 2 moves the radiation source 16 by rotating the arm unit 12 about the rotation axis 11. The console 2 also irradiates the breast M, which is the subject, with radiation under predetermined imaging conditions for tomosynthesis imaging at a plurality of radiation source positions obtained by moving the radiation source 16. The console 2 also obtains a plurality of projection images Gi (i=1 to n; n is the number of radiation source positions, for example, n=15) at a plurality of radiation source positions obtained by detecting radiation that has passed through the breast M with the radiation detector 15.

[0030] 5, the radiation source 16 is moved to each of the radiation source positions Si (i = 1 to n), and the radiation source 16 is driven at each radiation source position to irradiate the breast M with radiation. The radiation detector 15 detects the radiation that has passed through the breast M, and projection images G1, G2, ... Gn are acquired corresponding to each of the radiation source positions S1 to Sn. Note that the same dose of radiation is irradiated to the breast M at each of the radiation source positions S1 to Sn.

[0031] 5, the radiation source position Sc is the radiation source position where the optical axis X0 of the radiation emitted from the radiation source 16 is perpendicular to the detection surface 15A of the radiation detector 15. Hereinafter, the radiation source position Sc is referred to as the reference radiation source position Sc.

[0032] The console 2 reconstructs the multiple projection images Gi to generate multiple tomographic images in which desired slice planes of the breast M are emphasized. Specifically, the console 2 reconstructs the multiple projection images Gi using a well-known back projection method such as a simple back projection method or a filtered back projection method. As a result, as shown in FIG. 6, the console 2 generates multiple tomographic images Dj (j=1 to m) each representing a multiple slice plane of the breast M. At this time, a three-dimensional coordinate position in a three-dimensional space including the breast M is set, and pixel values ​​of pixels corresponding to the set three-dimensional coordinate position in the multiple projection images Gi are reconstructed, and the pixel value of the coordinate position is calculated.

[0033] The console 2 transfers the generated tomographic images Dj to an image processing device 4 or to an image storage system 3.

[0034] As an example, as shown in FIG. 7, the detection unit 42 detects regions including spicule candidates (hereinafter referred to as spicule candidate regions) from each of the multiple tomographic images Dj acquired by the acquisition unit 40. Spicules are radial line structures (hereinafter referred to as radial structures) that occur when breast cancer involves surrounding tissues during its development. The detection unit 42 detects regions including radially spreading line structures as spicule candidate regions from each of the tomographic images Dj. Note that normal mammary glands may appear in tomographic images as radial structures similar to spicules. For this reason, the spicule candidate regions may include normal mammary glands.

[0035] For the sake of simplicity, Fig. 7 illustrates a case in which the number of tomographic images Dj is five (i.e., j = 1 to 5). In the example illustrated in Fig. 7, tomographic image D2 includes spicule candidate region K21, tomographic image D3 includes spicule candidate region K31 and spicule candidate region K32, tomographic image D4 includes spicule candidate region K41 and spicule candidate region K42, and tomographic image D5 includes spicule candidate region K51.

[0036] The detection unit 42 detects spicule candidate regions from each of the multiple tomographic images Dj using a known spicule detection algorithm of computer-aided diagnosis (CAD). In the CAD spicule detection algorithm, the probability (likelihood) that a pixel in the tomographic image Dj is a spicule candidate region is derived, and a pixel with this probability equal to or greater than a predetermined threshold is detected as a spicule candidate region. Since it is difficult to distinguish spicules from radially spreading normal mammary glands using only one tomographic image Dj, radially spreading normal mammary glands are also detected as spicule candidate regions.

[0037] The detection process of the spicule candidate region is not limited to the one using CAD. For example, the detection process of the spicule candidate region may be a filtering process using a filter (e.g., a radial filter) to detect the spicule candidate region from the tomographic image Dj. The detection process of the spicule candidate region may be a detection model that has been machine-learned by deep learning or the like to detect the spicule candidate region from the tomographic image Dj.

[0038] The selection unit 44 selects a group of tomographic images corresponding to a plurality of spicule candidate regions that are presumed to represent the same spicule structure (i.e., radial structure) from among the plurality of spicule candidate regions detected by the detection unit 42. For example, the selection unit 44 selects a group of tomographic images corresponding to a plurality of spicule candidate regions that are captured in a plurality of successive tomographic images Dj in the thickness direction of the breast and that have the same planar position in each of the tomographic images Dj. "Almost the same planar position in the tomographic images Dj" means, for example, that at least a portion of the spicule candidate regions overlap between the tomographic images when projected onto a plane parallel to the detection surface 15A.

[0039] In the example shown in Fig. 7, the selection unit 44 infers that the spicule candidate regions K21, K31, and K41 and the spicule candidate regions K32, K42, and K51 each represent the same spicule structure. Specifically, as shown in Fig. 8, the selection unit 44 selects the tomographic images D2 to D4 as the first tomographic image group P1 because the spicule candidate regions K21, K31, and K41 appear in the successive tomographic images D2 to D4 and are located at approximately the same planar positions in the respective tomographic images D2 to D4. The selection unit 44 also selects the tomographic images D2 to D4 as the second tomographic image group P2 because the spicule candidate regions K32, K42, and K51 appear in the successive tomographic images D3 to D5 and are located at approximately the same planar positions in the respective tomographic images D3 to D5.

[0040] In the following description, a region including spicule candidate regions K21, K31, and K41 is referred to as a first region R1. A region including spicule candidate regions K32, K42, and K51 is referred to as a second region R2. The first region R1 and the second region R2 are, for example, rectangular regions.

[0041] The synthesis unit 46 generates a synthetic two-dimensional image SG using the group of tomographic images selected by the selection unit 44. The synthetic two-dimensional image is a pseudo two-dimensional image equivalent to a simple two-dimensional image captured by irradiating the breast M with radiation from the reference radiation source position Sc (see FIG. 5). In this embodiment, the synthesis unit 46 generates a synthetic two-dimensional image by synthesizing pixel values ​​of corresponding pixels in each of the tomographic images Dj in a stacked state as shown in FIG. 9 along the viewpoint direction from the reference radiation source position Sc toward the radiation detector 15, i.e., along the optical axis X0 (see FIG. 5). A specific example of the synthetic two-dimensional image generation process will be described below.

[0042] 9, in this embodiment, for the first region R1, the tomographic images D2-D4 included in the first tomographic image group P1 are synthesized, and for the second region R2, the tomographic images D3-D5 included in the second tomographic image group P2 are synthesized. For regions other than the first region R1 and the second region R2, all the tomographic images D1-D5 are synthesized. For example, the synthesis unit 46 sets the average value of pixel values ​​at corresponding pixel positions in the tomographic image groups to be synthesized as the pixel value of the synthesized two-dimensional image SG.

[0043] The determination unit 48 determines whether or not a spicule candidate region is a spicule, based on the composite two-dimensional image SG generated by the composition unit 46. As shown in Fig. 10, when a spicule candidate region in the composite two-dimensional image SG has a radial structure, the determination unit 48 determines the spicule candidate region to be a "spicule." On the other hand, when a spicule candidate region in the composite two-dimensional image SG does not have a radial structure, the determination unit 48 determines the spicule candidate region to be a "normal mammary gland."

[0044] Spicules are radial structures that arise when breast cancer involves surrounding tissues during its growth, and therefore lines radiate from the nucleus of breast cancer in all cross-sectional images included in the group of cross-sectional images to be synthesized. In other words, the central positions of the radial structures of spicules are almost the same between cross-sectional images. Therefore, spicules also have a radial structure centered on the nucleus in the synthesized two-dimensional image SG. In contrast, normal breast glands, which are similar to spicules, have radially spreading lines, but unlike breast cancer, they do not have a nucleus, so the central position of the radial structure changes between cross-sectional images. Therefore, the radial structure of normal breast glands disappears in the synthesized two-dimensional image SG, and the structure becomes non-radial. The determination unit 48 makes the above determination using this feature.

[0045] In the example shown in Fig. 10, the first region R1 is determined to be a "spicule" because a radial structure is detected. The second region R2 is determined to be a "normal mammary gland" because no radial structure is detected.

[0046] The determination unit 48 detects and determines the radial structure using CAD, similarly to the detection unit 42. For the detection and determination process of the radial structure, it is also possible to use a filtering process using a filter (e.g., a radial filter) or a detection model that has been machine-learned by deep learning or the like.

[0047] The display control unit 50 controls the display of the composite two-dimensional image SG generated by the composition unit 46 on the display 23.

[0048] Furthermore, the display control unit 50 performs control to display the determination result of the spicule candidate region by the determination unit 48 on the display 23. For example, the display control unit 50 performs control to display the determination result on the composite two-dimensional image SG. FIG. 11 shows an example of the display state of this determination result. As shown in FIG. 11, the display control unit 50 performs control to display a character string representing the determination result by the determination unit 48 near the spicule candidate region being determined. In the example shown in FIG. 11, the character string "spicula" is displayed near the spicule candidate region that has been determined to be a spicule. Note that a spicule candidate region with no character string displayed near it means that it is a "normal mammary gland".

[0049] Next, the operation of the image processing device 4 according to this embodiment will be described with reference to Fig. 12. The CPU 20 executes the image processing program 30 to execute the composite 2D image generation process shown in Fig. 12. The composite 2D image generation process shown in Fig. 12 is executed, for example, when a command to start execution is input by the user via the input device 24.

[0050] In step S10, the acquisition unit 40 acquires a plurality of tomographic images Dj generated by the console 2 causing the mammography apparatus 1 to perform tomosynthesis imaging.

[0051] In step S11, the detection unit 42 detects spicule candidate regions from each of the multiple tomographic images Dj acquired in step S10, as described above (see FIG. 7).

[0052] In step S12, the selection unit 44 selects a group of tomographic images corresponding to the spicule candidate regions detected from the same position in the tomographic image Dj from among the multiple spicule candidate regions detected by the detection unit 42 in step S11, as described above (see FIG. 8).

[0053] In step S13, the synthesis unit 46 generates a synthetic two-dimensional image SG using the group of tomographic images selected by the selection unit 44, as described above (see FIGS. 8 and 9).

[0054] In step S14, the determination unit 48 analyzes the composite two-dimensional image SG generated by the composition unit 46, as described above, to determine whether or not the spicule candidate region is a spicule (see FIG. 10).

[0055] In step S15, the display control unit 50 performs control to display the composite two-dimensional image SG generated by the composition unit 46 on the display 23, as described above.

[0056] In step S16, the display control unit 50 performs control to display the result of the determination of the spicule candidate region by the determination unit 48 on the display 23, as described above (see FIG. 11). When the process of step S16 ends, the composite two-dimensional image generation process ends.

[0057] As described above, in the composite 2D image generation process of this example, a composite 2D image SG is generated by combining a group of tomographic images corresponding to a plurality of spicule candidate regions detected from the same planar position in the tomographic image Dj. Therefore, in the composite 2D image SG, the spicule candidate regions will be radial if they are spicules and non-radial if they are normal mammary glands, so that the presence or absence of spicules can be determined with high accuracy.

[0058] As described above, according to this embodiment, spicula candidate regions are detected from a tomographic image, and therefore it is possible to generate a composite two-dimensional image and detect spicules in a short processing time compared to the conventional method of detecting candidates for lesions and the like from a three-dimensional image (3D volume data). That is, according to this embodiment, it is possible to accurately detect spicules in a short processing time. This reduces the burden of interpretation since a diagnostician such as a doctor only needs to interpret one composite two-dimensional image instead of a three-dimensional image. In addition, by deleting from the storage device tomographic images that are no longer necessary for interpretation, it is possible to save storage device capacity.

[0059] In the above embodiment, the display control unit 50 controls the display of the determination result by the determination unit 48 using text, but as shown in Fig. 13, the display control unit 50 may also control the display of the determination result by the determination unit 48 by making the color of the spicule candidate region indicating the determination result different from the color of the other regions. In the example shown in Fig. 13, the spicule candidate region determined to be a spicule is displayed with a dashed line. The dashed line indicates that the color is different from the other regions.

[0060] 14, the display control unit 50 may perform control to display a mark 60 indicating the determination result by the determination unit 48 near the spicule candidate region indicating the determination result. In the example shown in Fig. 14, a triangular mark 60 is displayed near the spicule candidate region determined to be a spicule.

[0061] Furthermore, in the above embodiment, for regions (first region R1 and second region R2 in the above embodiment) including a plurality of spicule candidate regions having the same position in the tomographic image Dj, the synthesis unit 46 sets the average of pixel values ​​at corresponding pixel positions in the group of tomographic images as the pixel value of the synthetic two-dimensional image SG. Alternatively, the synthesis unit 46 may set the average of pixel values ​​at corresponding pixel positions in a plurality of spicule candidate regions having the same position in the tomographic image Dj as the pixel value of the synthetic two-dimensional image SG. During this synthesis, for a pixel at a pixel position included in only one of the plurality of spicule candidate regions, the synthesis unit 46 sets the pixel value of that pixel as the pixel value of the synthetic two-dimensional image SG.

[0062] In the above embodiment, the synthesis unit 46 synthesizes all of the tomographic images included in the group of tomographic images selected by the selection unit 44, but it may synthesize some of the tomographic images included in the group of tomographic images. For example, in the example shown in Fig. 8, the synthesis unit 46 may synthesize only the tomographic image D2 and the tomographic image D4 among the tomographic images D2 to D4 included in the first tomographic image group P1. That is, the synthesis process by the synthesis unit 46 is not limited to synthesizing a plurality of continuous tomographic images, and may synthesize a plurality of discontinuous tomographic images.

[0063] In the above embodiment, the detection unit 42 detects regions having a radial structure from the tomographic images as spicule candidate regions, but in addition to the spicule candidate regions having a radial structure, regions having structures other than a radial structure may be detected as lesion candidate regions. In this case, it is preferable that the synthesis unit 46 change the number of tomographic images to be used for synthesis from the group of tomographic images selected by the selection unit 44 according to the type of candidate region.

[0064] Specifically, when the candidate region is not a spicule candidate region, the synthesis unit 46 selects fewer tomographic images from the group of tomographic images to perform synthesis processing than when the candidate region is a spicule candidate region. For example, when the candidate region is a spicule candidate region, the synthesis unit 46 performs synthesis processing using all or a plurality of tomographic images included in the group of tomographic images, and when the candidate region is a mass candidate region, the synthesis unit 46 performs synthesis processing using a part (preferably one) of the tomographic images included in the group of tomographic images. For example, in a mass candidate region, even if all the tomographic images included in the group of tomographic images are synthesized, the effect of eliminating the mass-like normal breast malformation erroneously detected by the detection unit 42 cannot be expected. Furthermore, since there is a possibility that the fine structure of the edge of the mass, which is important for a doctor to determine whether it is benign or malignant, may be hidden, it is preferable to perform synthesis processing using fewer tomographic images (preferably one included in the group of tomographic images) than the number of images used in the spicule candidate region.

[0065] In the above embodiment, the pixel value of the pixel of the composite two-dimensional image SG is the average value of the pixel values ​​of the corresponding pixels of the selected tomographic images. However, the present invention is not limited to this. The pixel value of the pixel of the composite two-dimensional image SG may be the weighted average value, the median value, the maximum value, or the minimum value of the pixel values ​​of the corresponding pixels of the selected tomographic images. Alternatively, a tomographic image having a pixel in which the difference between the pixel value of the pixel of interest and the average value of all tomographic images of the tomographic images is greater than a set value, a tomographic image having a pixel in which the variance value of the pixel values ​​of the region of interest including the pixel of interest is greater than the others, or a tomographic image having a pixel detected using edge detection may be selectively synthesized.

[0066] In the above embodiment, the determination unit 48 determines the spicula candidate region based on the composite two-dimensional image SG, but the determination unit 48 is not essential and may not be provided. In this case, a diagnostician such as a doctor may perform the determination based on the composite two-dimensional image SG displayed on the display 23.

[0067] In the above embodiment, the following various processors can be used as the hardware structure of the processing units that execute various processes, such as the acquisition unit 40, the detection unit 42, the selection unit 44, the synthesis unit 46, the determination unit 48, and the display control unit 50. As described above, the above various processors include a CPU, which is a general-purpose processor that executes software (programs) and functions as various processing units, as well as a programmable logic device (PLD), which is a processor whose circuit configuration can be changed after manufacture, such as an FPGA (Field Programmable Gate Array), a dedicated electric circuit, which is a processor having a circuit configuration designed specifically for executing specific processes, such as an ASIC (Application Specific Integrated Circuit), etc.

[0068] A single processing unit may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, multiple processing units may be configured with a single processor.

[0069] As an example of configuring multiple processing units with one processor, first, there is a form in which one processor is configured with a combination of one or more CPUs and software, as typified by computers such as client and server, and this processor functions as multiple processing units. Second, there is a form in which a processor is used that realizes the functions of the entire system including multiple processing units with one IC (Integrated Circuit) chip, as typified by System on Chip (SoC), etc. In this way, the various processing units are configured using one or more of the above various processors as a hardware structure.

[0070] Furthermore, more specifically, the hardware structure of these various processors can be an electric circuit that combines circuit elements such as semiconductor elements.

[0071] In the above embodiment, the image processing program 30 is stored (installed) in advance in the storage unit 22, but the present invention is not limited to this. The image processing program 30 may be provided in a form recorded on a recording medium such as a CD-ROM (Compact Disc Read Only Memory), a DVD-ROM (Digital Versatile Disc Read Only Memory), or a USB (Universal Serial Bus) memory. The image processing program 30 may be downloaded from an external device via a network. [Explanation of symbols]

[0072] 1. Mammography equipment 2 Console 3. Image storage system 4. Image Processing Device 11 Rotation axis 12 Arm section 13. Shooting stand 14 Radiation Unit 15 Radiation detector 15A Detection surface 16 Radiation source 17 Compression plate 18 Support part 19 Moving mechanism 20 CPU 21 Memory 22 Memory section 23 Display 24 Input Devices 25 Network I / F 27 Bus 30 Image Processing Programs 40 Acquisition Department 42 Detection unit 44 Selection section 46 Synthesis section 48 Judgment section 50 Display control unit 60 marks 100 Radiation imaging system Dj tomography Gi Projection Image K21, K31, K32, K41, K42, K51 Spicule candidate regions M Breast P1 First tomographic image group P2 Second tomographic image group R1 1st area R2 2nd area SG Composite 2D image Sc reference source position Si source position X0 optical axis

Claims

1. A method for detecting a spicule candidate region having a radial line structure from each of a plurality of tomographic images representing a plurality of tomographic planes of a subject, selecting, as a group of tomographic images, a plurality of the tomographic images corresponding to a plurality of the spicule candidate regions that represent the same spicule structure from among the detected plurality of spicule candidate regions; generating a composite two-dimensional image using the selected set of tomographic images; An image processing device including a processor, The processor, detects a lesion candidate region having a structure other than a radial line structure in addition to the spicule candidate region; if the candidate region is not the spicule candidate region, only one of the multiple tomographic images corresponding to a lesion candidate region having a structure other than a radial line structure is selected to generate the composite two-dimensional image. Image processing device.

2. The processor, determining whether the spicule candidate region is a spicule by analyzing the composite two-dimensional image; The image processing device according to claim 1 .

3. The processor, Controlling display of the composite two-dimensional image; and performing control to display a determination result as to whether or not the spicule candidate region is a spicule on the composite two-dimensional image; The image processing device according to claim 2 .

4. detecting a spicule candidate region having a radial line structure from each of a plurality of tomographic images representing a plurality of tomographic planes of the subject; selecting, as a group of tomographic images, a plurality of the tomographic images corresponding to a plurality of the spicule candidate regions that represent the same spicule structure from among the detected plurality of spicule candidate regions; generating a composite two-dimensional image using the selected set of tomographic images; An image processing method in which processing is performed by a processor included in an image processing device, detects a lesion candidate region having a structure other than a radial line structure in addition to the spicule candidate region; if the candidate region is not the spicule candidate region, only one of the multiple tomographic images corresponding to a lesion candidate region having a structure other than a radial line structure is selected to generate the composite two-dimensional image. The image processing method is performed by a processor.

5. detecting a spicule candidate region having a radial line structure from each of a plurality of tomographic images representing a plurality of tomographic planes of the subject; selecting, as a group of tomographic images, a plurality of the tomographic images corresponding to a plurality of the spicule candidate regions that represent the same spicule structure from among the detected plurality of spicule candidate regions; generating a composite two-dimensional image using the selected set of tomographic images; An image processing program for causing a processor of an image processing device to execute a process, detects a lesion candidate region having a structure other than a radial line structure in addition to the spicule candidate region; if the candidate region is not the spicule candidate region, only one of the multiple tomographic images corresponding to a lesion candidate region having a structure other than a radial line structure is selected to generate the composite two-dimensional image. An image processing program that causes the processor to execute processing.

Images