Radiation image processing apparatus, method, and program

Abstract

To accurately separate a plurality of components contained in a radiation image in a radiation image processing apparatus, method, and program.SOLUTION: A processor acquires first and second radiation images by radiation rays in which energy distributions are different of a subject containing a first component of a plurality of compositions and a second component of a single composition, derives an initial second component image in which the second component is highlighted on the basis of the first and second radiation images, specifies first and second component regions in the first and second radiation images on the basis of the initial second component image, derives characteristics of the first component on the attenuation of the radiation ray on the basis of the first and second radiation images in the first component region, derives characteristics of the first component in the second component region in the first or second radiation image on the basis of the characteristics of the first component in the first component region around the second component region, and derives first and second component images in which the first and second components are highlighted respectively on the basis of the characteristics of the first component at least in the subject region of the first or second radiation image.SELECTED DRAWING: Figure 3

Description

[Technical field]

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

[0002] Conventionally, energy subtraction processing has been known, which uses two radiation images obtained by irradiating a subject with two types of radiation with different energy distributions, utilizing the fact that the attenuation of transmitted radiation differs depending on the material that constitutes the subject. The energy subtraction processing is a method of matching the pixels of the two radiation images obtained as described above, multiplying the pixels by weighting coefficients based on attenuation coefficients according to the composition, and then performing subtraction (subtraction) to obtain an image in which specific compositions, such as bone and soft tissue, contained in the radiation image, are separated. In addition, a method has been proposed in which, in the energy subtraction processing, the thickness of the bone is derived from the acquired bone image and the thickness of the soft tissue is derived from the acquired soft tissue image, the weighting coefficients are updated based on the thickness of the bone and the thickness of the soft tissue, and the subtraction processing is performed using the updated weighting coefficients to update the soft tissue image and the bone image (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2021 / 054090 Summary of the Invention [Problem to be solved by the invention]

[0004] In order to separate the bones and soft parts of a subject by energy subtraction processing, the attenuation coefficient of the soft parts is necessary. Here, in a region in the radiographic image that contains only soft parts, the attenuation coefficient of the soft parts can be obtained directly from the radiographic image. However, in the bone region in the radiographic image, the soft parts and the bone parts overlap, so the attenuation coefficient of the soft parts cannot be obtained directly from the radiographic image. In addition, the soft parts are not composed of a single composition, but are a complex mixture of multiple compositions such as fat and muscle. In addition, the composition of the soft parts varies greatly from person to person. For this reason, in the bone region in the radiographic image where the soft parts and the bone parts overlap, it is difficult to accurately obtain the attenuation coefficient of the soft parts. In this way, if the attenuation coefficient of the soft parts cannot be obtained accurately, multiple components such as the bone parts and the soft parts cannot be accurately separated when processing is performed using the characteristics related to the attenuation of radiation, such as in the energy subtraction processing.

[0005] The present disclosure has been made in consideration of the above circumstances, and has an object to accurately separate multiple components contained in a radiological image. [Means for solving the problem]

[0006] A radiological image processing device according to the present disclosure includes at least one processor, the processor acquiring a first radiological image and a second radiological image by imaging a subject including a first component having a plurality of compositions and a second component having a single composition using radiation having different energy distributions; deriving an initial second component image in which the second component is enhanced based on the first radiation image and the second radiation image; identifying a second component region including the second component in the first radiographic image or the second radiographic image based on the initial second component image; identifying a region other than the second component region in the first radiographic image or the second radiographic image as a first component region including only the first component; deriving a characteristic of the first component related to attenuation of the radiation based on the first radiographic image and the second radiographic image in the identified first component region in the first radiographic image or the second radiographic image; deriving a characteristic of the first component in the identified second component region in the first radiological image or the second radiological image based on the characteristic of the first component derived in the first component region surrounding the second component region; A first component image in which the first component is emphasized and a second component image in which the second component is emphasized are derived based on characteristics of the first component derived in the first component region and the second component region, respectively, in at least the region of the subject in the first radiological image or the second radiological image.

[0007] In addition, in the radiation image processing device according to the present disclosure, the processor may update the initial second component image with the second component image, and repeat the steps of identifying the first component region using the updated initial second component image, identifying the second component region, deriving characteristics of the first component in the identified first component region, deriving characteristics of the first component in the identified second component region, and deriving the first component image and the second component image until a predetermined condition is satisfied.

[0008] In the radiological image processing device according to the present disclosure, the processor derives attenuation characteristics related to attenuation of the radiation in at least a region of the subject in the first radiological image or the second radiological image; The attenuation characteristic derived for the first component region may be used as a characteristic of the first component for the first component region.

[0009] In the radiological image processing device according to the present disclosure, a processor derives a first attenuation image and a second attenuation image representing an amount of attenuation of the radiation by the subject from the first radiological image and the second radiological image, respectively; An attenuation ratio, which is a ratio between corresponding pixels of the first attenuation image and the second attenuation image, may be derived as the characteristic of the first component.

[0010] In the radiological image processing device according to the present disclosure, a processor derives a first attenuation image and a second attenuation image representing an amount of attenuation of the radiation by the subject from the first radiological image and the second radiological image, respectively; An attenuation ratio, which is a ratio between corresponding pixels of the first attenuation image and the second attenuation image, may be derived as the attenuation characteristic.

[0011] In the radiation image processing device according to the present disclosure, a processor derives an initial second component attenuation image in which the second component is enhanced based on the first attenuation image, the second attenuation image, and a characteristic of the first component; deriving a second component attenuation image by matching a contrast of the second component included in the initial second component attenuation image with a contrast of the second component included in the first attenuation image or the second attenuation image; deriving a first component attenuation image in which the first component is emphasized based on the first attenuation image or the second attenuation image and the second component attenuation image; The first component image and the second component image may be derived from the first component attenuation image and the second component attenuation image, respectively.

[0012] In the radiological image processing device according to the present disclosure, the processor derives information regarding thicknesses of the plurality of components in at least a region of the subject in the first radiological image or the second radiological image; The attenuation coefficient of the first component may be derived as a characteristic of the first component based on information regarding the attenuation coefficient of the radiation of each of the plurality of compositions contained in the first component and the thickness of each of the plurality of compositions.

[0013] In the radiological image processing device according to the present disclosure, a processor derives a first attenuation image and a second attenuation image representing an amount of attenuation of the radiation by the subject from the first radiological image and the second radiological image, respectively; deriving a thickness of the first component and a thickness of the second component based on the first attenuation image, the second attenuation image, a characteristic of the first component, and a characteristic of the second component; The first component image and the second component image may be derived based on a thickness of the first component and a thickness of the second component.

[0014] In addition, in the radiological image processing device according to the present disclosure, the processor identifies a tentative first component region and a tentative second component region in the first radiological image or the second radiological image based on pixel values ​​of the first radiological image or the second radiological image; deriving a tentative first component characteristic relating to attenuation of the radiation in the tentative first component region of the first radiographic image or the second radiographic image based on the first radiographic image and the second radiographic image; deriving a property of a provisional first component in the provisional second component region of the first radiological image or the second radiological image based on a property of the provisional first component derived in the provisional first component region surrounding the provisional second component region; The initial second component image may be derived based on a characteristic of the provisional first component in at least a region of the subject in the first radiation image or the second radiation image.

[0015] In addition, in a radiation image processing device according to the present disclosure, the first component may be a soft part of the subject, the first component image may be a soft part image, the second component may be a bone part of the subject, the second component image may be a bone image, and the multiple components contained in the first component may be fat and muscle.

[0016] In the radiological image processing device according to the present disclosure, the processor may derive an evaluation result representing a bone condition based on a region of the bone in the bone image.

[0017] In the radiation image processing device according to the present disclosure, the evaluation result indicating the bone condition may be a bone salt amount or a value correlated with a bone salt amount.

[0018] In the radiological image processing device according to the present disclosure, the value correlated with bone salt mass may be a value representing the likelihood of osteoporosis.

[0019] In addition, in the radiation image processing device according to the present disclosure, the processor may determine a state of the bone based on a result of the evaluation.

[0020] In addition, in the radiation image processing device according to the present disclosure, the determination of the bone condition may be a determination of whether or not osteoporosis is suspected.

[0021] A radiation image processing method according to the present disclosure includes: acquiring a first radiation image and a second radiation image by imaging a subject including a first component having a plurality of compositions and a second component having a single composition using radiation having different energy distributions; deriving an initial second component image in which the second component is enhanced based on the first radiation image and the second radiation image; identifying a second component region including the second component in the first radiographic image or the second radiographic image based on the initial second component image; identifying a region other than the second component region in the first radiographic image or the second radiographic image as a first component region including only the first component; deriving a characteristic of the first component related to attenuation of the radiation based on the first radiographic image and the second radiographic image in the identified first component region in the first radiographic image or the second radiographic image; deriving a characteristic of the first component in the identified second component region in the first radiological image or the second radiological image based on the characteristic of the first component derived in the first component region surrounding the second component region; A first component image in which the first component is emphasized and a second component image in which the second component is emphasized are derived based on characteristics of the first component derived in the first component region and the second component region, respectively, in at least the region of the subject in the first radiological image or the second radiological image.

[0022] A radiological image processing program according to the present disclosure includes a step of acquiring a first radiological image and a second radiological image by imaging a subject including a first component having a plurality of compositions and a second component having a single composition using radiation having different energy distributions; deriving an initial second component image in which the second component is enhanced based on the first radiation image and the second radiation image; identifying a second component region including the second component in the first radiographic image or the second radiographic image based on the initial second component image; specifying a region other than the second component region in the first radiographic image or the second radiographic image as a first component region including only the first component; deriving a characteristic of the first component related to attenuation of the radiation in the identified first component region in the first radiographic image or the second radiographic image based on the first radiographic image and the second radiographic image; deriving a characteristic of the first component in the identified second component region in the first radiological image or the second radiological image based on a characteristic of the first component derived in the first component region surrounding the second component region; and deriving a first component image in which the first component is enhanced and a second component image in which the second component is enhanced based on characteristics of the first component derived in the first component region and the second component region, respectively, in at least the region of the subject in the first radiological image or the second radiological image. Effect of the Invention

[0023] According to the present disclosure, a plurality of components contained in a radiological image can be separated with high accuracy. [Brief description of the drawings]

[0024] [Figure 1] FIG. 1 is a schematic block diagram showing a configuration of a radiation image capturing system to which a radiation image processing device according to a first embodiment of the present disclosure is applied; [Diagram 2] FIG. 1 is a diagram showing a schematic configuration of a radiation image processing apparatus according to a first embodiment; [Diagram 3] FIG. 1 is a diagram showing a functional configuration of a radiation image processing apparatus according to a first embodiment; [Figure 4] FIG. 1 shows first and second radiation images. [Diagram 5] FIG. 2 is a diagram showing a schematic diagram of a process performed in the radiation image processing apparatus according to the first embodiment; [Figure 6] Diagram showing the display screen [Figure 7] 1 is a flowchart showing the processing performed in the first embodiment. [Figure 8] 1 is a flowchart showing the processing performed in the first embodiment. [Figure 9] FIG. 11 is a diagram illustrating a process performed in a radiation image processing apparatus according to a second embodiment. [Figure 10] FIG. 13 is a diagram illustrating a process performed in a radiation image processing apparatus according to a third embodiment. [Figure 11] Diagram showing fat and muscle thickness and attenuation coefficients of fat and muscle [Figure 12] FIG. 13 is a diagram showing the amount of attenuation depending on the thickness of bone and soft tissue in high-energy and low-energy images. [Figure 13] 11 is a flowchart showing the process performed in the third embodiment. [Figure 14] 11 is a flowchart showing the process performed in the third embodiment. [Figure 15] FIG. 13 is a diagram illustrating a process performed in a radiation image processing apparatus according to a fourth embodiment. [Figure 16]FIG. 13 is a diagram showing a functional configuration of a radiation image processing apparatus according to a fifth embodiment. [Figure 17] FIG. 13 is a diagram showing correction coefficients for correcting pixel values ​​of a bone image; [Figure 18] A diagram showing the bone mineral content display screen [Figure 19] A diagram showing an example of training data for training a trained model. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0025] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. Fig. 1 is a schematic block diagram showing the configuration of a radiographic image capturing system to which a radiographic image processing device according to a first embodiment of the present disclosure is applied. As shown in Fig. 1, the radiographic image capturing system according to the first embodiment includes a capturing device 1 and a radiographic image processing device 10 according to this embodiment.

[0026] The imaging device 1 is an imaging device for performing energy subtraction imaging by a so-called one-shot method in which radiation such as X-rays emitted from a radiation source 3 and transmitted through a subject H is irradiated to a first radiation detector 5 and a second radiation detector 6 with different energies. During imaging, as shown in Fig. 1, the first radiation detector 5, a radiation energy converting filter 7 made of a copper plate or the like, and the second radiation detector 6 are arranged in this order from the side closest to the radiation source 3, and the radiation source 3 is driven. The first and second radiation detectors 5 and 6 are in close contact with the radiation energy converting filter 7.

[0027] As a result, the first radiation detector 5 acquires a first radiation image G1 of the subject H using low-energy radiation including so-called soft rays. Meanwhile, the second radiation detector 6 acquires a second radiation image G2 of the subject H using high-energy radiation from which the soft rays have been removed. The first and second radiation images G1, G2 are input to the radiation image processing device 10.

[0028] The first and second radiation detectors 5, 6 are capable of repeatedly recording and reading out radiation images, and may be so-called direct type radiation detectors that generate electric charges by directly receiving radiation, or so-called indirect type radiation detectors that convert radiation into visible light and then convert the visible light into an electric charge signal. As a method for reading out the radiation image signal, it is preferable to use a so-called TFT readout method in which the radiation image signal is read out by turning on and off a TFT (thin film transistor) switch, or a so-called optical readout method in which the radiation image signal is read out by irradiating the detector with readout light, but this is not limiting and other methods may also be used.

[0029] Next, a radiological image processing device according to the first embodiment will be described. First, referring to FIG. 2, a hardware configuration of the radiological image processing device according to the first embodiment will be described. As shown in FIG. 2, the radiological image processing device 10 is a computer such as a workstation, a server computer, or a personal computer, and includes a CPU (Central Processing Unit) 11, a non-volatile storage 13, and a memory 16 as a temporary storage area. The radiological image processing device 10 also includes a display 14 such as a liquid crystal display, an input device 15 such as a keyboard and a mouse, and a network I / F (InterFace) 17 connected to a network (not shown). The CPU 11, the storage 13, the display 14, the input device 15, the memory 16, and the network I / F 17 are connected to a bus 18. The CPU 11 is an example of a processor in the present disclosure.

[0030] The storage 13 is realized by a hard disk drive (HDD), a solid state drive (SSD), a flash memory, etc. The storage 13 as a storage medium stores the radiation image processing program 12 installed in the radiation image processing device 10. The CPU 11 reads out the radiation image processing program 12 from the storage 13, expands it in the memory 16, and executes the expanded radiation image processing program 12.

[0031] The radiation image processing program 12 is stored in a storage device of a server computer connected to a network or in a network storage in an externally accessible state, and is downloaded and installed in the computer constituting the radiation image processing apparatus 10 upon request. Alternatively, the program is recorded on a recording medium such as a DVD (Digital Versatile Disc) or a CD-ROM (Compact Disc Read Only Memory) and distributed, and is installed in the computer constituting the radiation image processing apparatus 10 from the recording medium.

[0032] Next, the functional configuration of the radiation image processing device according to the first embodiment will be described. Fig. 3 is a diagram showing the functional configuration of the radiation image processing device according to the first embodiment. As shown in Fig. 3, the radiation image processing device 10 includes an image acquisition section 21, an area specification section 22, a characteristic derivation section 23, an image derivation section 24, a control section 25, and a display control section 26. The CPU 11 executes the radiation image processing program 12 to function as the image acquisition section 21, the area specification section 22, the characteristic derivation section 23, the image derivation section 24, the control section 25, and the display control section 26.

[0033] The image acquiring unit 21 acquires a first radiographic image G1 and a second radiographic image G2 of the subject H from the first and second radiation detectors 5 and 6 by making the imaging device 1 perform energy subtraction imaging of the subject H. At this time, imaging conditions such as imaging dose, energy distribution, tube voltage, SID (Source Image Receptor Distance) which is the distance between the radiation source 3 and the surfaces of the first and second radiation detectors 5 and 6, and SOD (Source Object Distance) which is the distance between the radiation source 3 and the surface of the subject H are set. The imaging conditions may be set by inputting from the input device 15 by the user. The set imaging conditions are stored in the storage 13. Note that the first and second radiographic images G1 and G2 may be acquired by a program separate from the radiographic image processing program according to the first embodiment. In this case, the image acquiring unit 21 reads out the first and second radiographic images G1 and G2 stored in the storage 13 from the storage 13 for processing.

[0034] FIG. 4 is a diagram showing the first and second radiation images. As shown in FIG. 4, the first and second radiation images G1, G2 include a region of the subject H and a direct radiation region obtained by directly irradiating the radiation detectors 5, 6 with radiation. The region of the subject H includes a soft tissue region and a bone region. The soft tissue components of the human body include muscle, fat, blood, and water. In this embodiment, non-fatty tissue including blood and water is treated as muscle. The soft tissue component and the bone component of the subject H are examples of the first component and the second component, respectively, of the present disclosure. Muscle and fat are examples of multiple compositions of the present disclosure.

[0035] The soft tissue regions of the first and second radiographic images G1, G2 contain only the soft tissue components of the subject H. The bone regions of the first and second radiographic images G1, G2 are actually regions where bone components and soft tissue components are mixed. The soft tissue regions are an example of a first component region containing only the first component according to the present disclosure, and the bone regions are an example of a second component region containing the second component according to the present disclosure.

[0036] The region identification unit 22, the characteristic derivation unit 23, the image derivation unit 24 and the control unit 25 will be described below together with the processing performed in the radiation image processing device according to the first embodiment. Fig. 5 is a diagram showing a schematic diagram of the processing performed in the radiation image processing device according to the first embodiment. In Fig. 5, for the sake of simplicity, the first radiation image G1 and the second radiation image G2 do not include a direct radiation region, and a rectangular bone region is included in the soft tissue region.

[0037] In the first embodiment, the region identifying unit 22 identifies bone regions and soft tissue regions in the first radiographic image G1 or the second radiographic image G2. To this end, the region identifying unit 22 derives attenuation characteristics related to attenuation of radiation in at least the region of the subject H in the first radiographic image G1 or the second radiographic image G2, and identifies the soft tissue regions and bone regions based on the attenuation characteristics in the region of the subject H. In the first embodiment, the region identifying unit 22 derives a first attenuation image CL and a second attenuation image CH representing the amount of attenuation of radiation by the subject H from the first radiographic image G1 and the second radiographic image G2, respectively, and derives an attenuation ratio, which is a ratio between corresponding pixels of the first attenuation image CL and the second attenuation image CH, as the attenuation characteristics.

[0038] Here, the pixel values ​​of the first attenuation image CL represent the amount of attenuation of low-energy radiation caused by the subject H, and the pixel values ​​of the second attenuation image CH represent the amount of attenuation of high-energy radiation caused by the subject H. The first attenuation image CL and the second attenuation image CH are derived from the first radiographic image G1 and the second radiographic image G2 by the following formulas (1) and (2). In formula (1), Gd1 is the pixel value of the direct radiation region in the first radiographic image G1, and in formula (2), Gd2 is the pixel value of the direct radiation region in the second radiographic image G2. CL(x,y)=Gd1-G1(x,y) (1) CH(x,y)=Gd2-G2(x,y) (2)

[0039] Next, the region specifying unit 22 derives an attenuation ratio map that indicates the attenuation ratio of radiation between the first radiographic image G1 and the second radiographic image G2. Specifically, the attenuation ratio map M1 is derived by deriving the ratio between corresponding pixels in the first attenuation image CL and the second attenuation image CH using the following formula (3). The attenuation ratio is an example of a characteristic related to the attenuation of radiation disclosed herein. M1(x,y)=CL(x,y) / CH(x,y) (3)

[0040] Here, in the first radiographic image G1 and the second radiographic image G2, the attenuation ratio of the region including only soft tissue components is smaller than the attenuation ratio of the region including bone components. Therefore, the region identifying unit 22 compares the attenuation ratio of each pixel in the attenuation ratio map M1, identifies a region consisting of pixels having an attenuation ratio larger than a predetermined threshold as a bone region, and identifies a region other than the bone region as a soft tissue region. The region identifying unit 22 may also compare the attenuation ratio of each pixel in the attenuation ratio map M1 with its surrounding pixels, and identify a pixel having a larger attenuation ratio than the surrounding pixels as a pixel of the bone region.

[0041] The property derivation unit 23 derives a property of the first component related to attenuation of radiation in a first component region including only the first component of the first radiographic image G1 or the second radiographic image G2, i.e., the soft tissue region, based on the first radiographic image G1 and the second radiographic image G2. The property derivation unit 23 also derives a property of the first component in the second component region, i.e., the bone region, based on the property of the first component derived in the soft tissue region surrounding the bone region. In this embodiment, the property derivation unit 23 derives the attenuation ratio between the first radiographic image G1 and the second radiographic image G2 as the property of the first component.

[0042] In the first embodiment, the region specifying unit 22 derives the attenuation ratio, which is a characteristic related to the attenuation of radiation, in the region of the subject H in the first radiographic image G1 or the second radiographic image G2. Therefore, the characteristic derivation unit 23 uses the attenuation ratio of the soft tissue region in the attenuation ratio map M1 derived by the region specifying unit 22 as a characteristic of the first component in the soft tissue region. That is, in the first embodiment, the region specifying unit 22 also functions as the characteristic derivation unit 23.

[0043] On the other hand, for the bone region, the attenuation ratio of the first component is derived by interpolating the attenuation ratios of the soft tissue regions surrounding the bone region. Instead of the interpolation, the attenuation ratio of the bone region may be derived as the median or average value of the attenuation ratios of the soft tissue regions in the attenuation ratio map M1 or a value that is a predetermined ratio from the side with a small attenuation ratio.

[0044] Here, when identifying the bone region, there is a possibility that a region that is erroneously detected or not detected may occur at the boundary between the bone region and the soft tissue region. For this reason, it is preferable that the region surrounding the bone region is a region that is within a range of, for example, 5 mm to 30 mm from the boundary of the identified bone region. In this case, for example, the average value of the soft tissue region surrounding the bone region can be used as the attenuation ratio of the soft tissue region surrounding the bone region. Note that instead of the average value, a median value, a maximum value, a minimum value, or the like may be used.

[0045] As a result, the characteristic derivation unit 23 derives a first component characteristic for the region of the subject H in the first radiographic image G1 or the second radiographic image G2. In the first embodiment, the characteristic of the first component is the attenuation ratio of the soft tissue region. In the first embodiment, the derived characteristic of the first component is used as a soft tissue erasure coefficient K1 for erasing soft tissue when the image derivation unit 24, which will be described later, derives a bone image.

[0046] The image derivation unit 24 derives a first component image in which the first component is emphasized and a second component image in which the second component is emphasized based on the characteristics of the first component derived in each of the soft tissue region and the bone region in the region of the subject H in the first radiological image G1 or the second radiological image G2. Specifically, the image derivation unit 24 derives a soft tissue image Gs in which the soft tissue component is emphasized and a bone image Gb in which the bone component is emphasized.

[0047] In the first embodiment, the image derivation unit 24 first derives an initial second component attenuation image in which the second component is emphasized, that is, an initial bone attenuation image in which the bone is emphasized, based on the first attenuation image CL, the second attenuation image CH, and the soft part elimination coefficient K1, which is a characteristic of the first component. Specifically, the image derivation unit 24 derives an initial bone attenuation image Cb0 by the following formula (4). Cb0(x,y)=CL(x,y)-CH(x,y)×K1(x,y) (4)

[0048] Here, the pixel value of the bone region of the initial bone attenuation image Cb0 derived as described above represents the difference between the attenuation amount of the bone replaced with the attenuation amount of the soft part by assuming that there is a soft part corresponding to the thickness of the bone, and the actual attenuation amount of the bone. Therefore, the image has a lower contrast than the bone attenuation image that is originally desired to be derived. If such a bone attenuation image with a low contrast is used, it is not possible to remove the bone components well when the soft attenuation image is derived by subtracting the bone attenuation image from the first attenuation image CL or the second attenuation image CH as described later.

[0049] In the first embodiment, the image derivation unit 24 derives the bone attenuation image Cb1 by matching the contrast of the initial bone attenuation image Cb0 with the contrast of the first attenuation image CL or the second attenuation image CH. In the first embodiment, the contrast of the initial bone attenuation image Cb0 is matched with the contrast of the first attenuation image CL. For this reason, the image derivation unit 24 multiplies the initial bone attenuation image Cb0 by a contrast conversion coefficient to convert the contrast of the initial bone attenuation image Cb0. Then, the image derivation unit 24 derives the correlation between the initial bone attenuation image Cb0 and a difference image ΔCG derived by subtracting the initial bone attenuation image Cb0 after the contrast conversion from the first attenuation image CL. Then, the contrast conversion coefficient is determined so that the correlation is minimized, and the bone attenuation image Cb1 is derived by multiplying the initial bone attenuation image Cb0 by the determined contrast conversion coefficient.

[0050] A table showing a contrast conversion coefficient that associates the contrast of the initial bone attenuation image Cb0 with the body thickness may be created in advance. In this case, the body thickness of the subject H may be derived by measuring or the like, a contrast conversion coefficient may be derived from the contrast and body thickness of the initial bone attenuation image Cb0 by referring to the table, and the initial bone attenuation image Cb0 may be converted by the derived contrast conversion coefficient to derive the bone attenuation image Cb1.

[0051] Then, the image derivation unit 24 derives the soft tissue attenuation image Cs1 by subtracting the bone attenuation image Cb1 from the first attenuation image CL according to the following formula (5). Cs1(x,y)=CL(x,y)-Cb1(x,y) (5)

[0052] Furthermore, the image derivation unit 24 derives a bone image Gb and a soft tissue image Gs by the following equations (6) and (7). Gb(x,y) = Gd1(x,y) - Cb1(x,y) (6) Gs(x,y)=Gd2(x,y)-Cs1(x,y) (7)

[0053] The control unit 25 controls the region identifying unit 22 to identify the bone region and the soft tissue region in the first radiographic image G1 and the second radiographic image G2 based on the second component image, i.e., the bone image Gb. Here, in the bone image, the bone region has a higher luminance than the soft tissue region and therefore has a smaller pixel value. Therefore, the region identifying unit 22 compares the pixel value of each pixel of the bone image Gb with a predetermined threshold value, identifies a region consisting of pixels whose pixel value is less than the predetermined threshold value as the bone region, and identifies a region other than the bone region as the soft tissue region. Note that the bone region may be derived from the bone image Gb using a trained model that has been machine-learned to extract the bone region from the bone image Gb.

[0054] Next, the control unit 25 controls the characteristic derivation unit 23 to derive the characteristic of the first component, i.e., the soft tissue erasure coefficient K1, for the area of ​​the subject H in the first radiographic image G1 or the second radiographic image G2, in the same manner as described above, using the bone area and soft tissue area identified by the area identification unit 22 based on the bone image Gb.

[0055] Furthermore, the control unit 25 derives a first component image in which the first component is emphasized and a second component image in which the second component is emphasized based on the characteristics of the first component in the region of the subject H in the first radiographic image G1 or the second radiographic image G2 in the same manner as described above. If the soft tissue image and bone image derived as described above are denoted by reference characters Gs1 and Gb1, respectively, the image derivation unit 24 derives a soft tissue image Gs2 in which the soft tissue component is emphasized and a bone image Gb2 in which the bone component is emphasized. Note that the soft tissue image and bone image derived by the image derivation unit 24 for the kth time are denoted by reference characters Gsk and Gbk, respectively.

[0056] The control unit 25 repeats the process of identifying the bone region and the soft region using the derived bone image Gbk, deriving the characteristics of the first component, and deriving the soft tissue image Gsk+1 and the bone image Gbk+1 until a predetermined condition is satisfied. The predetermined condition may be, for example, that a representative value of the difference values ​​between corresponding pixel values ​​of the bone image Gbk and the bone image Gbk+1 is less than a threshold value, but is not limited to this. Note that the representative value may be the sum of the absolute values ​​of the difference values, the average value of the absolute values ​​of the difference values, or the like.

[0057] The display control unit 26 displays the bone image Gb and the soft tissue image Gs on the display 14. Fig. 6 is a diagram showing a display screen for the bone image Gb and the soft tissue image Gs. As shown in Fig. 6, the bone image Gb and the soft tissue image Gs are displayed on the display screen 30.

[0058] Next, the process performed in the first embodiment will be described. FIG. 7 and FIG. 8 are flowcharts showing the process performed in the first embodiment. First, the image acquisition unit 21 causes the imaging device 1 to perform energy subtraction imaging of the subject H to acquire the first and second radiographic images G1 and G2 (radiographic image acquisition: step ST1). Next, the region identification unit 22 derives the first attenuation image CL and the second attenuation image CH from the first radiographic image G1 and the second radiographic image G2 (attenuation image derivation: step ST2), and identifies a soft tissue region containing only soft tissue components and a bone region containing bone in the first attenuation image CL or the second attenuation image CH (step ST3). Next, the characteristic derivation unit 23 derives the characteristic (attenuation ratio) of the soft tissue component related to the attenuation of the radiographic image in the soft tissue region (step ST4). Next, the characteristic derivation unit 23 derives the characteristic of the soft tissue component in the bone region (step ST5).

[0059] Next, the image derivation unit 24 derives an initial bone attenuation image Cb0 (step ST6), and converts the contrast of the initial bone attenuation image Cb0 to derive a bone attenuation image Cb1 (step ST7). Furthermore, the image derivation unit 24 subtracts the bone attenuation image Cb1 from the first attenuation image CL to derive a soft tissue attenuation image Cs1 (step ST8). Next, the image derivation unit 24 derives a bone image Gb and a soft tissue image Gs by the above-mentioned formulas (6) and (7) (step ST9).

[0060] Next, the control unit 25 judges whether the bone image Gb and the soft tissue image Gs are derived for the first time (step ST10). If step ST10 is negative, the control unit 25 judges whether the bone image Gbk derived this time and the bone image Gbk-1 derived last time satisfy a predetermined condition (step ST11). If step ST10 is positive and if step ST11 is negative, the control unit 25 controls the region specification unit 22 to specify the soft tissue region and the bone region in the first and second radiographic images G1 and G2 based on the bone image Gb (step ST12). After step ST12, the process returns to step ST4, and the processes from step ST4 onward are repeated. If step ST11 is positive, the display control unit 26 displays the bone image Gb and the soft tissue image Gs (step ST13), and the process ends.

[0061] In this way, in the first embodiment, after deriving the bone image Gb and the soft tissue image Gs, the bone region and the soft tissue region are specified based on the bone image Gb, and further, the attenuation ratio in the bone region including the bone components of the first radiographic image G1 or the second radiographic image G2 is derived based on the attenuation ratio of the soft tissue components derived in the soft tissue region surrounding the bone region. Then, based on the attenuation ratio in at least the region of the subject H in the first radiographic image G1 or the second radiographic image G2, the soft tissue image Gs in which the soft tissue components are emphasized and the bone image Gb in which the bone components are emphasized are derived. Furthermore, these processes are repeated until a predetermined condition is satisfied.

[0062] Here, since the bone image Gb is an image in which the bone regions in the first radiographic image G1 and the second radiographic image G2 are emphasized, the bone region can be identified more accurately than the bone region identified in the first processing, i.e., based on the attenuation ratio map M1. This improves the accuracy of separation between the bone region and the soft tissue region, and therefore the characteristics of the first component in the soft tissue region and the characteristics of the first component in the bone region can be obtained with high accuracy. Therefore, the attenuation ratio of the soft tissue component in the bone region can be derived with high accuracy, and as a result, the soft tissue image Gs and the bone image Gb in which the soft tissue component and the bone component are separated with high accuracy can be derived.

[0063] In the first embodiment, the first attenuation image CL and the second attenuation image CH representing the amount of attenuation of radiation are derived from the first radiographic image G1 and the second radiographic image G2, and the soft tissue image Gs and the bone image Gb are derived using the first attenuation image CL and the second attenuation image CH. Therefore, when the derived amount of attenuation is used as the soft tissue erasure coefficient K1 for erasing the soft tissue component, the soft tissue component can be effectively erased by the above formula (4). Therefore, the soft tissue image Gs and the bone image Gb in which the soft tissue component and the bone component are accurately separated can be derived.

[0064] Next, a second embodiment of the present disclosure will be described. Note that the functional configuration of the radiation image processing device according to the second embodiment is the same as the functional configuration of the radiation image processing device according to the first embodiment shown in Fig. 3, so a detailed description of the functional configuration will be omitted. In the second embodiment, the process performed by the region identification unit 22 is different from that of the first embodiment.

[0065] FIG. 9 is a diagram showing a schematic diagram of the process performed in the second embodiment. As shown in FIG. 9, in the first process of deriving the bone image Gb and the soft tissue image Gs, the region specifying unit 22 detects the bone region from the first attenuation image CL or the second attenuation image CH. The bone region may be detected from the first radiographic image G1 or the second radiographic image G2. For this purpose, in the second embodiment, the region specifying unit 22 uses a trained model constructed by machine learning a neural network so as to detect the bone region from the radiographic image or the attenuation image. In this case, the trained model is constructed so as to detect the bone region by learning the bone region based on the pixel values ​​of the radiographic image or the attenuation image.

[0066] On the other hand, in a radiographic image or attenuation image, the pixel values ​​of the bone region are significantly different from those of the soft region containing only soft components. Therefore, the radiographic image or attenuation image may be subjected to threshold processing to detect the bone region from the radiographic image or attenuation image. Also, in a radiographic image or attenuation image, the shape of the bone region is specified by the difference between the pixel values ​​of the bone region and the pixel values ​​of the soft region containing only soft components. Therefore, the bone region may be detected from the radiographic image or attenuation image by template matching using the shape of the bone region according to the part of the subject H contained in the radiographic image or attenuation image.

[0067] In the second embodiment, the characteristic derivation unit 23 derives the attenuation ratio of the soft tissue component in the bone region identified based on the pixel values ​​of the radiographic image or attenuation image by the region identification unit 22 as described above. The process after deriving the attenuation ratio of the soft tissue component in the bone region is the same as in the first embodiment, and therefore a detailed description thereof will be omitted here.

[0068] In the second embodiment, when the region specifying unit 22 specifies the bone region and the soft tissue region from the first radiographic image G1 or the second radiographic image G2, the characteristic derivation unit 23 may derive the first attenuation image CL and the second attenuation image CH. In this case, the characteristic derivation unit 23 may derive the attenuation ratio of the soft tissue region, the attenuation ratio of the soft tissue component for the bone region, and the attenuation ratio map M1.

[0069] In the first and second embodiments, the initial bone attenuation image Cb0, the bone attenuation image Cb1, and the soft tissue attenuation image Cs1 are derived using the first attenuation image CL, the second attenuation image CH, and the soft tissue erasure coefficient K1, and then the bone image Gb and the soft tissue image Gs are derived, but this is not limited to this. The bone image Gb and the soft tissue image Gs may be derived by the following formulas (8) and (9) using the first radiation image G1, the second radiation image G2, and the soft tissue erasure coefficient K1. Gb(x,y)=G1(x,y)-K1(x,y)×G2(x,y) (8) Gs(x,y)=G1(x,y)-Gb(x,y) (9)

[0070] Next, a third embodiment of the present disclosure will be described. Note that the functional configuration of the radiation image processing device according to the third embodiment is the same as the functional configuration of the radiation image processing device according to the first embodiment shown in Fig. 3, so detailed description of the functional configuration will be omitted. In the first embodiment, the attenuation ratio of the soft part component is derived as the characteristic of the first component, but in the third embodiment, the soft part attenuation coefficient, which is the attenuation coefficient of the soft part component of low energy radiation and high energy radiation, is derived as the characteristic of the first component, which is different from the first embodiment.

[0071] 10 is a diagram showing a schematic diagram of the process performed in the third embodiment. In the third embodiment, the process until the region specifying unit 22 derives the first attenuation image CL and the second attenuation image CH is the same as that in the first and second embodiments, so a detailed description is omitted here.

[0072] In the third embodiment, in the first process of deriving the bone image Gb and the soft tissue image Gs, the region specifying unit 22 derives the fat ratio at each pixel position of the first and second radiographic images G1, G2 using the attenuation coefficients of fat and muscle for high-energy radiation and low-energy radiation, respectively. Then, the region specifying unit 22 specifies the bone region and the soft tissue region based on the derived fat ratio.

[0073] Here, the amount of attenuation of radiation by subject H depends on the thickness and radiation quality (high energy or low energy) of soft tissue and bone. For this reason, if the attenuation coefficient representing the attenuation rate per unit thickness is μ, the amount of attenuation of radiation CL0, CH0 at each pixel position in the low-energy image and high-energy image, respectively, can be expressed by the following formulas (10) and (11). In formulas (10) and (11), ts is the thickness of soft tissue, tb is the thickness of bone, μLs is the soft-tissue attenuation coefficient of low-energy radiation, μLb is the bone attenuation coefficient of low-energy radiation, μHS is the soft-tissue attenuation coefficient of high-energy radiation, and μHB is the bone attenuation coefficient of high-energy radiation. CL0=μLs(ts,tb)×ts+μLb(ts,tb)×tb (10) CH0=μHs(ts,tb)×ts+μHb(ts,tb)×tb (11)

[0074] In equations (10) and (11), the amount of attenuation CL0 of the low-energy image corresponds to the pixel value of the first attenuation image CL, and the amount of attenuation CH0 of the high-energy image corresponds to the pixel value of the second attenuation image CH. Therefore, equations (10) and (11) are expressed by the following equations (12) and (13). Note that equations (10) to (13) all express the relationship between the pixels of the first attenuation image CL and the second attenuation image CH, but omit (x, y) indicating the pixel position. CL=μLs(ts,tb)×ts+μLb(ts,tb)×tb (12) CH=μHs(ts,tb)×ts+μHb(ts,tb)×tb (13)

[0075] The soft tissue thickness ts and the bone thickness tb can be derived by solving equations (12) and (13) using the soft tissue thickness ts and the bone thickness tb as variables. In order to solve equations (12) and (13), the soft tissue attenuation coefficients μLs, μHs and the bone attenuation coefficients μLb, μHb for low-energy radiation and high-energy radiation, respectively, are required. Here, since there is no difference in composition of the bone depending on the subject H, the bone attenuation coefficients μLb, μHb depending on the soft tissue thickness ts and the bone thickness tb can be prepared in advance.

[0076] On the other hand, soft tissue cannot be prepared in advance because muscle and fat are mixed in a complex manner and the ratio of muscle and fat varies depending on the subject H. For this reason, in the third embodiment, the region specifying unit 22 derives the soft tissue attenuation coefficient μLs for low-energy radiation and the soft tissue attenuation coefficient μHs for high-energy radiation using the first attenuation image CL and the second attenuation image CH. The derivation of the soft tissue attenuation coefficients μLs and μHs will be described below.

[0077] In the third embodiment, the soft-tissue attenuation coefficient is derived on the premise that the composition constituting the soft tissue is the highest density composition of muscle and the lowest density composition of fat, and the mixed composition of fat and muscle has an intermediate value between the attenuation coefficients of both. First, the region specifying unit 22 calculates provisional soft-tissue attenuation coefficients μ0Ls and μ0Hs for low-energy radiation and high-energy radiation, respectively, by setting the fat ratio at each pixel position to N%, and performing weighted addition of the fat attenuation coefficient and the muscle attenuation coefficient in a ratio of N:100-N while sequentially increasing N from 0. Note that when fat and muscle overlap, the attenuation coefficient changes due to the influence of the radiation hardening of the component (usually fat) present on the radiation source 3 side, but in this embodiment, the influence of the radiation hardening is not taken into consideration. For this reason, the fat ratio N% used in this embodiment does not match the actual body fat percentage of the subject H. The processing is based on the assumption that the actual soft tissue attenuation coefficient is a value between the attenuation coefficient of fat and the attenuation coefficient of muscle shown in Figure 11. In Figure 11, the horizontal axis represents the thickness (mm) of fat and muscle, and the vertical axis represents the attenuation coefficient.

[0078] Next, the region specifying unit 22 calculates the body thickness TN when the fat ratio is N% from the pixel value of the first attenuation image CL and the provisional soft-tissue attenuation coefficient μ0Ls for the low-energy image by the following formula (14). At this time, the body thickness TN is calculated assuming that pixels including bones are also composed of only soft tissue. TN(x,y)=CL(x,y) / μ0Ls(x,y) (14)

[0079] Next, the region specifying unit 22 calculates the attenuation amount CHN1 of the high-energy radiation using the body thickness TN calculated using formula (14) and the provisional soft tissue attenuation coefficient μHs for the high-energy radiation using the following formula (15). Then, the region specifying unit 22 calculates a difference value ΔCH by subtracting the second attenuation image CH from the attenuation amount CHN1 using the following formula (16). CHN1(x,y)=TN(x,y)×μ0Hs(x,y) (15) ΔCH(x,y)=CHN1(x,y)-CH(x,y) (16)

[0080] When the difference value ΔCH is negative, the provisional soft part attenuation coefficients μ0Ls and μ0Hs are smaller than the correct soft part attenuation coefficients, i.e., closer to fat. When the difference value ΔCH is positive, the provisional soft part attenuation coefficients μ0Ls and μ0Hs are closer to muscle. The region specifying unit 22 calculates the provisional soft part attenuation coefficients μ0Ls and μ0Hs for all pixels of the first attenuation image CL and the second attenuation image CH while changing N so that the difference value ΔCH approaches 0. Then, when the difference value ΔCH becomes 0 or a predetermined threshold value or less, N is determined as the fat ratio for that pixel. In addition, the region specifying unit 22 determines the provisional soft part attenuation coefficients μ0Ls and μ0Hs when calculating the determined fat ratio N as the soft part attenuation coefficients μLs and μHs. When the difference value ΔCH is negative, the fat ratio N is increased, and when the difference value ΔCH is positive, the fat ratio N is decreased.

[0081] Here, in the first radiographic image G1 and the second radiographic image G2, the fat percentage is a value close to 0 or a negative value in pixels of an area including bone components. If the subject H is a human, the fat percentage cannot be 0 or a negative value. For this reason, the region identifying unit 22 identifies, as bone regions in the first radiographic image G1 and the second radiographic image G2, regions consisting of pixels in which the fat percentage N is a value close to 0 (for example, a value less than a predetermined threshold value) or a negative value. In addition, the region identifying unit 22 identifies, as soft tissue regions, regions other than bone regions in the first radiographic image G1 and the second radiographic image G2.

[0082] In the third embodiment, the characteristic derivation unit 23 derives the soft part attenuation coefficients μLs, μHs as the characteristics of the first component. Here, in the third embodiment, the region identification unit 22 derives the soft part attenuation coefficients μLs, μHs for each pixel of the first radiographic image G1 or the second radiographic image G2. Therefore, the characteristic derivation unit 23 uses the soft part attenuation coefficients μLs, μHs derived by the region identification unit 22 as the characteristics of the first component in the soft part region. That is, in the third embodiment, the region identification unit 22 also has a function as the characteristic derivation unit 23.

[0083] On the other hand, the characteristic derivation unit 23 derives the soft tissue attenuation coefficients μLs, μHs in the bone region by interpolating the soft tissue attenuation coefficients of the soft tissue regions surrounding the bone region. Instead of the interpolation, the soft tissue attenuation coefficients μLs, μHs in the soft tissue region may be derived as the median, average, or a value that is a predetermined ratio from the side of the smaller attenuation coefficient as the soft tissue attenuation coefficients μLs, μHs for the bone region. In this way, the characteristic derivation unit 23 derives the characteristic of the first component for at least the region of the subject H in the first radiographic image G1 or the second radiographic image G2.

[0084] In the third embodiment, the image derivation unit 24 derives a soft tissue image Gs in which the soft tissue components are enhanced and a bone image Gb in which the bone components are enhanced. In the third embodiment, the image derivation unit 24 derives the soft tissue thickness ts and the bone thickness tb based on the soft tissue attenuation coefficients μLs, μHs derived by the characteristic derivation unit 23 and the bone attenuation coefficients μLb, μHb derived in advance, and derives the soft tissue image Gs and the bone image Gb based on the derived soft tissue thickness ts and bone thickness tb.

[0085] The above-mentioned formulas (12) and (13) are used to derive the thickness ts of the soft tissue and the thickness tb of the bone tissue. As described above, the image derivation unit 24 derives the thickness ts of the soft tissue and the thickness tb of the bone tissue by solving formulas (12) and (13) using the thickness ts of the soft tissue and the thickness tb of the bone tissue as variables. Note that the thickness ts of the soft tissue and the thickness tb of the bone tissue are derived at each pixel of the first attenuation image CL and the second attenuation image CH, but in the following description, (x, y) representing the pixel position is omitted.

[0086] The image derivation unit 24 first calculates the thickness ts0 of the soft part when the thickness of the bone part is set to tb=0 using equation (13). When tb=0 and ts=ts0, CH=μHs(ts,0)×ts0+μHb(ts0,0)×0=μHs(ts0,0)×ts0, so ts0 is calculated using the following equation (17). The image derivation unit 24 also calculates the thickness tb0 of the bone part when the thickness of the soft part is set to ts=0 using equation (13). When ts=0 and tb=tb0, CH=μHs(0,tb0)×0+μHb(0,tb0)×tb0, so tsb is calculated using the following equation (18). Note that (x,y) representing pixel position is omitted in equations (17) and (18). ts0=CH / μHS(ts0,0) (17) tb0 = CH / μHb(tb0,0) (18)

[0087] FIG. 12 is a diagram showing the relationship of the attenuation amount according to the thickness of the bone portion and the thickness of the soft portion. In FIG. 12, the attenuation amount 33 indicates the attenuation amount which is the pixel value of the low energy image and the attenuation amount which is the pixel value of the high energy image derived by actually photographing the subject. For the sake of explanation, in the attenuation amount 33, the attenuation amount which is the pixel value of the low energy image and the attenuation amount which is the pixel value of the high energy image are given the same symbols CL and CH as the first attenuation image and the second attenuation image, respectively. Here, the attenuation amount of the low energy image and the attenuation amount of the high energy image become larger as the composition becomes denser. Therefore, the composition when tb=0 and ts=ts0 is set has a lower density than the composition based on the actual thickness of the bone portion and the thickness of the soft portion. Therefore, when tb = 0 and ts = ts0, the pixel value, i.e., the amount of attenuation, of the first attenuation image (here referred to as a provisional first attenuation image CL') derived by equation (12) is smaller than the pixel value of the first attenuation image CL derived from the actual thickness of the bone and the thickness of the soft tissue, as shown by attenuation amount 34 in Figure 12 (i.e., CL > CL').

[0088] On the other hand, when tb=tb0 and ts=0, the composition becomes denser than the composition based on the actual thickness of the bone and the thickness of the soft tissue. Therefore, when tb=tb0 and ts=0, the pixel value of the provisional first attenuation image CL′ derived by the formula (12), i.e., the amount of attenuation, becomes smaller than the pixel value of the first attenuation image CL derived from the actual thickness of the bone and the thickness of the soft tissue, as shown by the amount of attenuation 35 in FIG. 12 (i.e., CL <CL′)。

[0089] The pixel values ​​of the provisional first attenuation image CL', i.e., the amount of attenuation, derived by equation (12) using the actual thickness of the bone and the thickness of the soft tissue, are the same as the pixel values ​​of the first attenuation image CL, as shown by the amount of attenuation 36 in Fig. 12. Using this, the image derivation unit 24 derives the thickness tb of the bone and the thickness ts of the soft tissue as follows.

[0090] (Step 1) First, in equation (13), the provisional soft tissue thickness tsk is calculated using the pixel value of the second attenuation image CH and the soft tissue attenuation coefficient μHs derived for each pixel. Note that the initial value of the provisional bone thickness tbk is set to 0.

[0091] (Step 2) Next, a provisional first attenuation image CL' is calculated according to the above formula (12) using the calculated provisional soft tissue thickness tsk, provisional bone thickness tbk, soft tissue attenuation coefficient μLs and bone attenuation coefficient μLb for low-energy radiation.

[0092] (Step 3) Next, a difference value ΔCL of each pixel between the provisional first attenuation image CL′ and the first attenuation image CL is calculated. The difference value ΔCL is assumed to be a pixel value of the amount of radiation attenuated by the bone, and the provisional thickness tbk of the bone is updated.

[0093] (Step 4) Next, a provisional second attenuation image CH' is calculated by the above formula (13) using the updated provisional bone thickness tbk and provisional soft tissue thickness tsk.

[0094] (Step 5) Next, a difference value ΔCH between the provisional second attenuation image CH′ and the provisional second attenuation image CH is calculated. The difference value ΔCH is assumed to be a pixel value of the amount of radiation attenuated by the soft tissue, and the provisional thickness tsk of the soft tissue is updated.

[0095] Then, the process of steps 1 to 5 is repeated until the absolute values ​​of the difference values ​​ΔCL and ΔCH become less than a predetermined threshold value, thereby deriving the thickness ts of the soft part and the thickness tb of the bone part. Note that the process of steps 1 to 5 may be repeated a predetermined number of times to derive the thickness ts of the soft part and the thickness tb of the bone part.

[0096] The image derivation unit 24 then derives a soft tissue image Gs based on the derived soft tissue thickness ts, and derives a bone image Gb based on the derived bone thickness tb. Here, the soft tissue image Gs has pixel values ​​whose size corresponds to the soft tissue thickness ts, and the bone image Gb has pixel values ​​whose size corresponds to the bone thickness Gb.

[0097] Furthermore, in the third embodiment, similarly to the first and second embodiments, the control unit 25 controls the region identification unit 22 to identify the bone region and the soft tissue region in the first radiographic image G1 and the second radiographic image G2 based on the second component image, i.e., the bone image Gb. Next, the control unit 25 derives the soft tissue attenuation coefficients μLs and μHs as the characteristics of the first component in each of the bone region and the soft tissue region identified by the region identification unit 22 based on the bone image Gb. Furthermore, similarly to the above, the control unit 25 derives the soft tissue thickness ts and the bone thickness tb based on the soft tissue attenuation coefficients μLs and μHs derived in each of the bone region and the soft tissue region and the bone attenuation coefficients μLb and μHb derived in advance, and derives the soft tissue image Gs2 and the bone image Gb2 based on the derived soft tissue thickness ts and the bone thickness tb.

[0098] The control unit 25 repeats the process of identifying the bone region and the soft region using the derived bone image Gbk, deriving the characteristics of the first component, and deriving the soft tissue image Gsk+1 and the bone image Gbk+1 until a predetermined condition is satisfied. The predetermined condition can be, for example, a condition that a representative value of the difference values ​​of corresponding pixel values ​​between the bone image Gbk and the bone image Gbk+1 is less than a threshold value, but is not limited to this. Note that, as the representative value, the sum of the absolute values ​​of the difference values, the average value of the absolute values ​​of the difference values, etc. can be used.

[0099] Next, the process performed in the third embodiment will be described. FIG. 13 and FIG. 14 are flowcharts showing the process performed in the third embodiment. First, the image acquisition unit 21 causes the imaging device 1 to perform energy subtraction imaging of the subject H to acquire the first and second radiographic images G1 and G2 (radiographic image acquisition: step ST21). Next, the region identification unit 22 derives the first attenuation image CL and the second attenuation image CH from the first radiographic image G1 and the second radiographic image G2 (attenuation image derivation: step ST22), and identifies a soft tissue region containing only soft tissue components and a bone region containing bone in the first attenuation image CL or the second attenuation image CH (step ST23). Next, the characteristic derivation unit 23 derives the characteristic of the soft tissue component related to the attenuation of the radiographic image in the soft tissue region (soft tissue attenuation coefficient) (step ST24). Next, the characteristic derivation unit 23 derives the characteristic of the soft tissue component in the bone region (step ST25).

[0100] Next, the image derivation unit 24 derives the soft tissue thickness ts and the bone thickness tb (step ST26), and derives the bone image Gb and the soft tissue image Gs from the soft tissue thickness ts and the bone thickness tb (step ST27).

[0101] Next, the control unit 25 judges whether the bone image Gb and the soft tissue image Gs are derived for the first time (step ST28). If step ST28 is negative, the control unit 25 judges whether the bone image Gbk derived this time and the bone image Gbk-1 derived last time satisfy a predetermined condition (step ST29). If step ST28 is positive and if step ST29 is negative, the control unit 25 controls the region specification unit 22 to specify the soft tissue region and the bone region based on the bone image Gb (step ST30). After step ST30, the process returns to step ST24, and the process from step ST24 onward is repeated. If step ST29 is positive, the display control unit 26 displays the bone image Gb and the soft tissue image Gs (step ST31), and the process ends.

[0102] In this way, in the third embodiment, in a bone region including bone components in the first radiographic image G1 or the second radiographic image G2, a soft tissue attenuation coefficient is derived based on the characteristics of the soft tissue components derived in the soft tissue region surrounding the bone region, i.e., the soft tissue attenuation coefficient. Then, a soft tissue image Gs in which the soft tissue components are enhanced and a bone image Gb in which the bone components are enhanced are derived based on the soft tissue attenuation coefficient in at least the region of the subject H in the first radiographic image G1 or the second radiographic image G2. Furthermore, these processes are repeated until a predetermined condition is satisfied.

[0103] Therefore, similar to the first embodiment, the soft tissue attenuation coefficient can be derived more accurately in the bone region, and as a result, a soft tissue image Gs and a bone image Gb in which the soft tissue components and the bone components are accurately separated can be derived.

[0104] Next, a fourth embodiment of the present disclosure will be described. The functional configuration of the radiological image processing device according to the fourth embodiment is the same as the functional configuration of the radiological image processing device according to the first embodiment shown in FIG. 3, so detailed description of the functional configuration will be omitted. In the fourth embodiment, the process performed by the region identification unit 22 is different from that of the third embodiment. FIG. 15 is a diagram showing a schematic diagram of the process performed in the fourth embodiment. As shown in FIG. 15, the region identification unit 22 detects a bone region from the first attenuation image CL or the second attenuation image CH. Note that the bone region may be detected from the first radiological image G1 or the second radiological image G2. For this reason, in the fourth embodiment, the region identification unit 22 uses a trained model constructed by machine learning a neural network so as to detect a bone region from a radiological image or an attenuation image, as in the second embodiment. In this case, the trained model is constructed so as to detect a bone region by learning the bone region based on the pixel values ​​of the radiological image or the attenuation image.

[0105] On the other hand, in a radiographic image or attenuation image, the pixel values ​​of the bone region are significantly different from those of the soft region containing only soft components. Therefore, the radiographic image or attenuation image may be subjected to threshold processing to detect the bone region from the radiographic image or attenuation image. Also, in a radiographic image or attenuation image, the shape of the bone region is specified by the difference between the pixel values ​​of the bone region and the pixel values ​​of the soft region containing only soft components. Therefore, the bone region may be detected from the radiographic image or attenuation image by template matching using the shape of the bone region according to the part of the subject H contained in the radiographic image or attenuation image.

[0106] In the fourth embodiment, the characteristic derivation unit 23 derives the soft tissue attenuation coefficient in the bone region identified by the region identification unit 22 based on the pixel values ​​of the radiographic image or attenuation image as described above. The process after deriving the soft tissue attenuation coefficient in the bone region is the same as in the third embodiment, and therefore a detailed description thereof will be omitted here.

[0107] In the fourth embodiment, when the region specifying unit 22 specifies the bone region and the soft tissue region from the first radiographic image G1 or the second radiographic image G2, the characteristic derivation unit 23 may derive the first attenuation image CL and the second attenuation image CH. In this case, the characteristic derivation unit 23 may derive the fat ratio and the soft tissue attenuation coefficient.

[0108] In the above third and fourth embodiments, the first attenuation image CL and the second attenuation image CH are used to derive the soft tissue attenuation coefficient to derive the bone image Gb and the soft tissue image Gs, but the present invention is not limited to this. The soft tissue attenuation coefficient may be derived from the first radiographic image G1 and the second radiographic image G2, and the thickness of the bone and the thickness of the soft tissue may be derived to derive the bone image Gb and the soft tissue image Gs.

[0109] Next, a fifth embodiment of the present disclosure will be described. Fig. 16 is a diagram showing a functional configuration of a radiological image processing device according to the fifth embodiment. In Fig. 16, the same components as those in Fig. 3 are given the same reference numerals, and detailed description will be omitted here. As shown in Fig. 16, a radiological image processing device 10A according to the fifth embodiment differs from the first to fourth embodiments in that it includes an evaluation unit 27 that derives an evaluation result representing a bone state based on a bone region in a bone image.

[0110] The evaluation unit 27 derives, for example, bone mineral content as an evaluation result representing the state of the bone. The derivation of bone mineral content will be described below. The evaluation unit 27 derives bone mineral content for each pixel of the bone image Gb. In the fifth embodiment, the evaluation unit 27 derives bone mineral content B by converting each pixel value of the bone image Gb into a pixel value of a bone image acquired under standard imaging conditions. Specifically, the evaluation unit 27 derives bone mineral content by correcting each pixel value of the bone image Gb using a correction coefficient acquired from a lookup table described later.

[0111] Here, the higher the tube voltage of the radiation source 3 and the higher the energy of the radiation emitted from the radiation source 3, the lower the contrast between the soft tissue and the bone in the radiographic image. In addition, in the process in which the radiation passes through the subject H, low-energy components of the radiation are absorbed by the subject H, causing beam hardening in which the radiation becomes high-energy. The higher the body thickness of the subject H, the greater the increase in energy of the radiation due to beam hardening. Specifically, the higher the tube voltage, the lower the contrast between the bone and the soft tissue relative to the body thickness of the subject H. In addition, when the body thickness of the subject H exceeds a certain value, the larger the body thickness, the lower the contrast. In addition, the higher the pixel value of the bone region in the bone image Gb, the higher the contrast between the bone and the soft tissue. For this reason, the higher the pixel value of the bone region in the bone image Gb, the higher the contrast shifts to the higher side.

[0112] In this embodiment, a lookup table for acquiring correction coefficients for correcting the difference in contrast in the bone image Gb according to the tube voltage at the time of imaging and the decrease in contrast due to the effect of beam hardening is stored in the storage 13 of the radiation image processing device 10. The correction coefficients are coefficients for correcting each pixel value of the bone image Gb.

[0113] FIG. 17 is a diagram showing an example of a lookup table for acquiring a correction coefficient. In FIG. 17, a lookup table (hereinafter, simply referred to as a table) LUT1 in which a reference imaging condition is set to a tube voltage of 90 kV is illustrated. As shown in FIG. 17, in the table LUT1, a larger correction coefficient is set as the tube voltage increases and as the body thickness of the subject H increases. In the example shown in FIG. 17, since the reference imaging condition is a tube voltage of 90 kV, when the tube voltage is 90 kV and the body thickness is 0, the correction coefficient is 1. Note that, although the table LUT1 is shown two-dimensionally in FIG. 17, the correction coefficient differs depending on the pixel value of the bone region. Therefore, the table LUT1 is actually a three-dimensional table to which an axis representing the pixel value of the bone region is added.

[0114] The evaluation unit 27 derives the body thickness (body thickness distribution) of the subject H for each pixel in the soft tissue image Gs. The body thickness can be derived by any method, such as the method described in JP 2015-043959 A. The body thickness distribution can be obtained by subtracting SOD from SID.

[0115] The evaluation unit 27 extracts from the table LUT1 a correction coefficient C0(x, y) for each pixel according to the imaging conditions including the body thickness distribution T(x, y) of the subject H and the set value of the tube voltage stored in the storage 13. Then, as shown in the following formula (19), the evaluation unit 27 multiplies each pixel (x, y) of the bone region in the bone image Gb by the correction coefficient C0(x, y) to obtain the bone mineral content B(x, y) (g / cm) for each pixel of the bone image Gb. 2The bone mineral content B(x, y) derived in this manner represents the pixel values ​​of the bone region contained in the radiographic image obtained by imaging the subject H under the standard imaging condition of a tube voltage of 90 kV and from which the effects of beam hardening have been removed. B(x,y)=C0(x,y)×Gb(x,y) (19)

[0116] In the fifth embodiment, the display control unit 26 displays the bone mineral content derived by the evaluation unit 27 on the display 14. Fig. 18 is a diagram showing a display screen of the bone mineral content. As shown in Fig. 18, a display screen 40 has a bone mineral content display area 41.

[0117] A bone image Gb is displayed in the bone mineral content display area 41. In the bone image Gb, a pattern is applied to the bone area according to the bone mineral content. In FIG. 18, for the sake of simplicity, a pattern representing the bone mineral content is applied only to the vertebrae including the fifth lumbar vertebra L5. A reference 42 indicating the magnitude of the bone mineral content for the applied pattern is displayed below the bone mineral content display area 41. The operator can easily recognize the bone mineral content by reading the bone image Gb while referring to the reference 42. Instead of a pattern, a different color may be applied to the bone image Gb according to the bone mineral content.

[0118] The evaluation unit 27 may derive a value correlated with bone mineral content as the evaluation result. An example of the value correlated with bone mineral content is an evaluation value representing the likelihood of osteoporosis. In this case, the evaluation unit 27 has a trained model that outputs an evaluation value representing the likelihood of osteoporosis when an image of a region of a target bone such as a lumbar vertebrae in the bone image Gb, i.e., a distribution of pixel values ​​of the target bone, is input. The trained model is constructed by machine learning a neural network using an image of a lumbar vertebrae (hereinafter referred to as an image of a normal lumbar vertebrae) extracted from a bone image of a patient without osteoporosis and an image of a lumbar vertebrae of an osteoporotic patient as training data. FIG. 19 is a diagram showing an example of training data for training the trained model. As shown in FIG. 19, first training data 51 representing a normal lumbar vertebrae and second training data 52 representing an osteoporotic lumbar vertebrae are used as training data for training the trained model. A normal lumbar vertebrae has a dense microstructure of bone trabeculae, whereas an osteoporotic lumbar vertebrae has a sparse microstructure of bone trabeculae compared to a normal lumbar vertebrae.

[0119] The neural network is machine-trained so that the output is 0 when first teacher data 51 is input, and the output is 1 when second teacher data 52 is input. In this way, a trained model is constructed so that when an input image of a lumbar spine indicates osteoporosis, an evaluation value close to 1 is output. In this case, the neural network learns the distribution of pixel values ​​of the lumbar spine image and the microstructure of the bone trabeculae, and a trained model is constructed so as to output an evaluation value representing the likelihood of osteoporosis from the distribution of pixel values ​​and the microstructure of the lumbar spine image.

[0120] Here, the evaluation unit 27 may use the evaluation value output by the trained model itself as the evaluation result, or may determine whether or not osteoporosis is suspected by comparing the evaluation value with a threshold value.

[0121] In the above-described embodiments, the first and second radiographic images G1, G2 are acquired by a one-shot method, but the present invention is not limited to this. The first and second radiographic images G1, G2 may be acquired by a so-called two-shot method, in which imaging is performed twice using only one radiation detector. In the case of the two-shot method, the position of the subject H included in the first radiographic image G1 and the second radiographic image G2 may be shifted due to the body movement of the subject H. For this reason, it is preferable to align the position of the subject in the first radiographic image G1 and the second radiographic image G2 before carrying out the processing of the present embodiment.

[0122] In the above embodiment, the radiation images are acquired in a system that captures an image of the subject H using the first and second radiation detectors 5 and 6. However, the technology of the present disclosure can be applied to a case where the first and second radiation images G1 and G2 are acquired using stimulable phosphor sheets instead of radiation detectors. In this case, two stimulable phosphor sheets are stacked and irradiated with radiation that has passed through the subject H, and the radiation image information of the subject H is accumulatively recorded on each stimulable phosphor sheet, and the radiation image information is photoelectrically read from each stimulable phosphor sheet to acquire the first and second radiation images G1 and G2. The two-shot method may also be used when the first and second radiation images G1 and G2 are acquired using stimulable phosphor sheets.

[0123] Moreover, the radiation in the above embodiment is not particularly limited, and in addition to X-rays, α-rays, γ-rays, etc. can be used.

[0124] In the above embodiment, the following various processors can be used as the hardware structure of the processing unit that executes various processes, such as the image acquisition unit 21, the area identification unit 22, the characteristic derivation unit 23, the image derivation unit 24, the control unit 25, the display control unit 26, and the evaluation unit 27. 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), and 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).

[0125] 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 (e.g., 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.

[0126] 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.

[0127] Furthermore, more specifically, the hardware structure of these various processors can be an electric circuit that combines circuit elements such as semiconductor elements.

[0128] The following are appended claims to this disclosure. (Additional note 1) at least one processor; The processor, A first radiographic image and a second radiographic image are obtained by imaging a subject including a first component having a plurality of compositions and a second component having a single composition using radiation having different energy distributions; deriving an initial second component image in which the second component is enhanced based on the first radiation image and the second radiation image; identifying a second component region including the second component in the first radiographic image or the second radiographic image based on the initial second component image; identifying a region other than the second component region in the first radiographic image or the second radiographic image as a first component region including only the first component; deriving a characteristic of the first component related to attenuation of the radiation based on the first radiographic image and the second radiographic image in the identified first component region in the first radiographic image or the second radiographic image; deriving a characteristic of the first component in the identified second component region in the first radiological image or the second radiological image based on the characteristic of the first component derived in the first component region surrounding the second component region; a radiation image processing device that derives a first component image in which the first component is enhanced and a second component image in which the second component is enhanced based on characteristics of the first component derived in the first component region and the second component region, respectively, in at least the region of the subject in the first radiation image or the second radiation image. (Additional note 2) The radiation image processing device according to claim 1, wherein the processor updates the initial second component image with the second component image, and repeats the steps of identifying the first component region using the updated initial second component image, identifying the second component region, deriving characteristics of the first component in the identified first component region, deriving characteristics of the first component in the identified second component region, and deriving the first component image and the second component image, until a predetermined condition is satisfied. (Additional note 3) The processor derives attenuation characteristics regarding attenuation of the radiation in at least a region of the subject in the first radiographic image or the second radiographic image; 3. The radiation image processing device according to claim 1 or 2, wherein the attenuation characteristic derived for the first component region is used as a characteristic of the first component for the first component region. (Additional note 4) the processor derives a first attenuation image and a second attenuation image representing an amount of attenuation of the radiation by the subject from the first radiographic image and the second radiographic image, respectively; 4. The radiological image processing device according to claim 1, further comprising: a ratio between corresponding pixels of the first attenuation image and the second attenuation image, the ratio being an attenuation ratio, which is a ratio between corresponding pixels of the first attenuation image and the second attenuation image, being derived as a characteristic of the first component. (Additional note 5) the processor derives a first attenuation image and a second attenuation image representing an amount of attenuation of the radiation by the subject from the first radiographic image and the second radiographic image, respectively; The radiation image processing device according to claim 3, wherein an attenuation ratio, which is a ratio between corresponding pixels of the first attenuation image and the second attenuation image, is derived as the attenuation characteristic. (Additional note 6) The processor derives an initial second component attenuation image in which the second component is enhanced based on the first attenuation image, the second attenuation image, and characteristics of the first component; deriving a second component attenuation image by matching a contrast of the second component included in the initial second component attenuation image with a contrast of the second component included in the first attenuation image or the second attenuation image; deriving a first component attenuation image in which the first component is emphasized based on the first attenuation image or the second attenuation image and the second component attenuation image; 5. The radiation image processing device according to claim 4, wherein the first component image and the second component image are derived from the first component attenuation image and the second component attenuation image, respectively. (Additional note 7) the processor derives information regarding thicknesses of the plurality of components in at least a region of the object in the first radiographic image or the second radiographic image; 3. The radiological image processing device according to claim 1 or 2, further comprising: a step of deriving an attenuation coefficient of the first component as a characteristic of the first component, based on information regarding the attenuation coefficient of the radiation of each of the plurality of compositions contained in the first component and a thickness of each of the plurality of compositions. (Additional note 8) the processor derives a first attenuation image and a second attenuation image representing an amount of attenuation of the radiation by the subject from the first radiographic image and the second radiographic image, respectively; Deriving a thickness of the first component and a thickness of the second component based on the first attenuation image, the second attenuation image, a characteristic of the first component, and a characteristic of the second component; 8. The radiation image processing device according to claim 7, wherein the first component image and the second component image are derived based on a thickness of the first component and a thickness of the second component. (Additional note 9) the processor identifies a tentative first component region and a tentative second component region in the first radiographic image or the second radiographic image based on pixel values ​​of the first radiographic image or the second radiographic image; deriving a provisional first component characteristic relating to attenuation of the radiation based on the first radiographic image and the second radiographic image in the provisional first component region of the first radiographic image or the second radiographic image; deriving a property of a provisional first component in the provisional second component region of the first radiological image or the second radiological image based on a property of the provisional first component derived in the provisional first component region surrounding the provisional second component region; 9. The radiological image processing device according to claim 1, further comprising: a processor for deriving an initial second component image based on a characteristic of the provisional first component in at least a region of the subject in the first radiological image or the second radiological image. (Additional note 10) 10. The radiation image processing device according to any one of claims 1 to 9, wherein the first component is a soft part of the subject, the first component image is a soft part image, the second component is a bone part of the subject, the second component image is a bone image, and the multiple components contained in the first component are fat and muscle. (Additional note 11) 11. The radiological image processing device according to claim 10, wherein the processor derives an evaluation result representing a state of the bone based on a region of the bone in the bone image. (Additional note 12) 12. The radiation image processing device according to claim 11, wherein the evaluation result indicating the bone condition is a bone mineral content or a value correlated with a bone mineral content. (Additional note 13) 13. The radiation image processing device according to claim 12, wherein the value correlated with the bone salt amount is a value representing a likelihood of osteoporosis. (Additional note 14) 14. The radiological image processing device according to claim 11, wherein the processor determines a state of the bone based on a result of the evaluation. (Additional note 15) 15. The radiation image processing device according to claim 14, wherein the determination of the bone condition is a determination of whether or not osteoporosis is suspected. (Additional note 16) a computer acquires a first radiographic image and a second radiographic image by imaging a subject including a first component having a plurality of compositions and a second component having a single composition using radiation having different energy distributions; deriving an initial second component image in which the second component is enhanced based on the first radiation image and the second radiation image; identifying a second component region including the second component in the first radiographic image or the second radiographic image based on the initial second component image; identifying a region other than the second component region in the first radiographic image or the second radiographic image as a first component region including only the first component; deriving a characteristic of the first component related to attenuation of the radiation based on the first radiographic image and the second radiographic image in the identified first component region in the first radiographic image or the second radiographic image; deriving a characteristic of the first component in the identified second component region in the first radiological image or the second radiological image based on the characteristic of the first component derived in the first component region surrounding the second component region; a first component image in which the first component is enhanced and a second component image in which the second component is enhanced based on characteristics of the first component derived in the first component region and the second component region, respectively, in at least the region of the subject in the first radiological image or the second radiological image. (Additional note 17) a step of acquiring a first radiographic image and a second radiographic image by imaging a subject including a first component having a plurality of compositions and a second component having a single composition using radiation having different energy distributions; deriving an initial second component image in which the second component is enhanced based on the first radiation image and the second radiation image; identifying a second component region including the second component in the first radiographic image or the second radiographic image based on the initial second component image; specifying a region other than the second component region in the first radiographic image or the second radiographic image as a first component region including only the first component; deriving a characteristic of the first component related to attenuation of the radiation in the identified first component region in the first radiographic image or the second radiographic image based on the first radiographic image and the second radiographic image; deriving a characteristic of the first component in the identified second component region in the first radiological image or the second radiological image based on a characteristic of the first component derived in the first component region surrounding the second component region; and deriving a first component image in which the first component is enhanced and a second component image in which the second component is enhanced based on characteristics of the first component derived in the first component region and the second component region, respectively, in at least the region of the subject in the first radiographic image or the second radiographic image. [Explanation of symbols]

[0129] 1. Imaging device 3 Radiation source 5, 6 Radiation detectors 7 Radiation Energy Conversion Filter 10 Image processing device 11 CPU 12 Image Processing Programs 13. Storage 14 Display 15 Input Devices 16 Memory 17 Network Interface 18 Bus 21 Image acquisition unit 22 Area identification part 23 Characteristics Derivation Section 24 Image extraction section 25 Control Unit 26 Display control section 27 Evaluation Section 30,40 display screen 33~36 Attenuation amount 41 Bone mineral content display area Gb Bone image Gs Soft tissue image H Subject LUT1 Lookup table

Claims

1. At least one processor; The processor, A first radiographic image and a second radiographic image are obtained by imaging a subject including a first component having a plurality of compositions and a second component having a single composition using radiation having different energy distributions; deriving an initial second component image in which the second component is enhanced based on the first radiation image and the second radiation image; identifying a second component region including the second component in the first radiographic image or the second radiographic image based on the initial second component image; identifying a region in the first radiographic image or the second radiographic image other than the second component region as a first component region including only the first component; deriving a characteristic of the first component related to attenuation of the radiation based on the first radiographic image and the second radiographic image in the identified first component region in the first radiographic image or the second radiographic image; deriving a characteristic of the first component in the identified second component region in the first radiological image or the second radiological image based on the characteristic of the first component derived in the first component region surrounding the second component region; and deriving a first component image in which the first component is enhanced and a second component image in which the second component is enhanced based on characteristics of the first component derived in the first component region and the second component region, respectively, in at least the region of the subject in the first radiological image or the second radiological image.

2. 2. The radiation image processing device according to claim 1, wherein the processor updates the initial second component image with the second component image, and repeats the steps of identifying the first component region using the updated initial second component image, identifying the second component region, deriving characteristics of the first component in the identified first component region, deriving characteristics of the first component in the identified second component region, and deriving the first component image and the second component image until a predetermined condition is satisfied.

3. The processor derives attenuation characteristics regarding attenuation of the radiation in at least a region of the subject in the first radiographic image or the second radiographic image; The radiation image processing apparatus according to claim 1 , wherein the attenuation characteristic derived for the first component region is used as a characteristic of the first component for the first component region.

4. the processor derives a first attenuation image and a second attenuation image representing an attenuation amount of the radiation by the subject from the first radiographic image and the second radiographic image, respectively; The radiation image processing apparatus according to claim 1 , further comprising: a ratio between corresponding pixels of the first attenuation image and the second attenuation image, the ratio being an attenuation ratio, which is derived as the characteristic of the first component.

5. the processor derives a first attenuation image and a second attenuation image representing an attenuation amount of the radiation by the subject from the first radiographic image and the second radiographic image, respectively; The radiation image processing apparatus according to claim 3 , wherein an attenuation ratio, which is a ratio between corresponding pixels of the first attenuation image and the second attenuation image, is derived as the attenuation characteristic.

6. The processor derives an initial second component attenuation image in which the second component is enhanced based on the first attenuation image, the second attenuation image, and characteristics of the first component; deriving a second component attenuation image by matching a contrast of the second component included in the initial second component attenuation image with a contrast of the second component included in the first attenuation image or the second attenuation image; deriving a first component attenuation image in which the first component is emphasized based on the first attenuation image or the second attenuation image and the second component attenuation image; The radiation image processing apparatus according to claim 4 , wherein the first component image and the second component image are derived from the first component attenuation image and the second component attenuation image, respectively.

7. the processor derives information regarding thicknesses of the plurality of components in at least a region of the object in the first radiographic image or the second radiographic image; 3. The radiation image processing device according to claim 1, further comprising: a radiation attenuation coefficient of the first component that is derived as a characteristic of the first component based on information regarding the attenuation coefficient of the radiation of each of the plurality of compositions contained in the first component and the thickness of each of the plurality of compositions.

8. the processor derives a first attenuation image and a second attenuation image representing an attenuation amount of the radiation by the subject from the first radiographic image and the second radiographic image, respectively; Deriving a thickness of the first component and a thickness of the second component based on the first attenuation image, the second attenuation image, a characteristic of the first component, and a characteristic of the second component; The radiation image processing apparatus according to claim 7 , wherein the first component image and the second component image are derived based on a thickness of the first component and a thickness of the second component.

9. the processor identifies a tentative first component region and a tentative second component region in the first radiographic image or the second radiographic image based on pixel values ​​of the first radiographic image or the second radiographic image; deriving a tentative first component characteristic relating to attenuation of the radiation in the tentative first component region of the first radiographic image or the second radiographic image based on the first radiographic image and the second radiographic image; deriving a characteristic of a provisional first component in the provisional second component region of the first radiological image or the second radiological image based on a characteristic of the provisional first component derived in the provisional first component region surrounding the provisional second component region; The radiation image processing apparatus according to claim 1 , wherein the initial second component image is derived based on a characteristic of the provisional first component in at least the region of the subject in the first radiation image or the second radiation image.

10. 3. The radiation image processing device according to claim 1, wherein the first component is a soft part of the subject, the first component image is a soft part image, the second component is a bone part of the subject, the second component image is a bone image, and the multiple components contained in the first component are fat and muscle.

11. The apparatus according to claim 10 , wherein the processor derives an evaluation result representing a bone condition based on a region of the bone in the bone image.

12. The radiation image processing apparatus according to claim 11 , wherein the evaluation result indicating the bone condition is a bone mineral content or a value correlated with a bone mineral content.

13. The radiation image processing apparatus according to claim 12 , wherein the value correlated with the bone salt amount is a value representing a likelihood of osteoporosis.

14. The radiation image processing apparatus according to claim 11 , wherein the processor determines a state of the bone based on the evaluation result.

15. The radiation image processing apparatus according to claim 14, wherein the determination of the bone condition is a determination of whether osteoporosis is suspected.

16. a computer acquires a first radiographic image and a second radiographic image by imaging a subject including a first component having a plurality of compositions and a second component having a single composition using radiation having different energy distributions; deriving an initial second component image in which the second component is enhanced based on the first radiation image and the second radiation image; identifying a second component region including the second component in the first radiographic image or the second radiographic image based on the initial second component image; identifying a region in the first radiographic image or the second radiographic image other than the second component region as a first component region including only the first component; deriving a characteristic of the first component related to attenuation of the radiation based on the first radiographic image and the second radiographic image in the identified first component region in the first radiographic image or the second radiographic image; deriving a characteristic of the first component in the identified second component region in the first radiological image or the second radiological image based on the characteristic of the first component derived in the first component region surrounding the second component region; a first component image in which the first component is enhanced and a second component image in which the second component is enhanced based on characteristics of the first component derived in the first component region and the second component region, respectively, in at least the region of the subject of the first radiological image or the second radiological image.

17. a step of acquiring a first radiographic image and a second radiographic image by imaging a subject including a first component having a plurality of compositions and a second component having a single composition using radiation having different energy distributions; deriving an initial second component image in which the second component is enhanced based on the first radiation image and the second radiation image; identifying a second component region including the second component in the first radiographic image or the second radiographic image based on the initial second component image; specifying a region other than the second component region in the first radiographic image or the second radiographic image as a first component region including only the first component; deriving a characteristic of the first component related to attenuation of the radiation in the identified first component region in the first radiographic image or the second radiographic image based on the first radiographic image and the second radiographic image; deriving a characteristic of the first component in the identified second component region in the first radiological image or the second radiological image based on a characteristic of the first component derived in the first component region surrounding the second component region; and deriving a first component image in which the first component is enhanced and a second component image in which the second component is enhanced based on characteristics of the first component derived in the first component region and the second component region, respectively, in at least the region of the subject of the first radiological image or the second radiological image.

Images