Radiation imaging processing device, method, and program

Abstract

In a radiation image processing device, method, and program, estimation processing using an estimation model is performed with high accuracy. [Solution] The processor acquires two radiographic images having contrast based on a first characteristic corresponding to a first shooting condition, derives the body thickness distribution of the subject based on at least one of the two radiographic images, removes scattered radiation components from the two radiographic images based on the first characteristic, derives a first characteristic soft tissue image and a first characteristic bone tissue image representing the soft tissue and bone tissue of the subject from the two radiographic images, converts the first characteristic soft tissue image and the first characteristic bone tissue image into a second characteristic soft tissue image and a second characteristic bone tissue image having contrast based on the second characteristic based on the first characteristic, a second characteristic corresponding to a second shooting condition, and body thickness distribution, and derives a processed target radiographic image having contrast based on the second characteristic by adding these together.

Description

Technical Field

[0005] ,

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

Background Art

[0002] Conventionally, when performing a diagnosis using a radiation image, a comparative reading using the patient's past radiation images has been performed. Here, when two radiation images for comparative reading are acquired by different imaging devices, the contrast between the two will be different. Also, if the imaging conditions (tube voltage of the radiation source, imaging distance, tube current, etc.) at the time of imaging are different, the contrast of the two radiation images will be different. Further, the contrast of a radiation image also changes depending on scattered radiation included in the radiation image. In view of such a situation, in order to perform comparative reading accurately, a technique for matching the contrast of two radiation images with different imaging conditions and characteristics of the acquired devices has been proposed (see, for example, Patent Document 1).

[0003] On the other hand, a computer-aided diagnosis system (CAD: Computer Aided Diagnosis, hereinafter referred to as CAD) that automatically detects structures such as abnormal shadows in a radiation image using a learned estimation model such as AI (Artificial Intelligence) and performs highlighting of the detected structures has also been proposed.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] As mentioned above, different shooting conditions result in different contrasts in the acquired radiation images. Therefore, when training an AI, it is preferable to use radiation images acquired under various shooting conditions as training data. However, acquiring radiation images under various shooting conditions is difficult from the perspective of human exposure. For this reason, it is difficult to perform estimation processing with estimation models with high accuracy.

[0006] This disclosure is made in view of the above circumstances and aims to enable high-precision estimation processing using estimation models. [Means for solving the problem]

[0007] The radiation image processing device according to this disclosure is a radiation image processing device that performs estimation processing on radiation images by using an estimation model constructed through learning, Equipped with a processor, The aforementioned processor, Two target radiation images with different energy distributions are obtained by photographing the subject under the first shooting conditions, and a first characteristic corresponding to the first shooting conditions is obtained. The second characteristic is obtained according to the second imaging conditions when acquiring the training radiation image for which the estimation model was constructed. Based on at least one of the two target radiation images, the thickness distribution of the subject is derived. The scattered radiation component, which is contained in the radiation that has passed through the subject and is scattered by the subject, is removed from the two target radiation images based on the first characteristic. By weighting and subtracting the two target radiation images from which the scattered radiation components have been removed, a first characteristic soft tissue image representing the soft tissue of the subject and a first characteristic bone tissue image representing the bone tissue of the subject are derived. Based on the first characteristic, the second characteristic, and the body thickness distribution, the first characteristic soft tissue image and the first characteristic bone image are converted into a second characteristic soft tissue image and a second characteristic bone image having contrast based on the second characteristic, By adding the second characteristic soft tissue image and the second characteristic bone image, a processed target radiographic image having contrast based on the second characteristic is derived. The processed target radiation image is input to the estimation model and the estimation process is performed.

[0008] In the radiation image processing apparatus according to this disclosure, the processor derives a scattered radiation component corresponding to the second characteristic based on the second characteristic and the thickness distribution, Furthermore, the processed target radiation image may be derived using the derived scattered radiation components.

[0009] In the radiation image processing apparatus according to the present disclosure, the first characteristic includes at least one of the following: the energy of the radiation according to the first shooting conditions; a radiation attenuation coefficient according to the thickness distribution of an object interposed between the subject and a radiation detector that detects the radiation transmitted through the subject when acquiring the two target radiation images; a ratio of the scattered radiation component contained in the radiation transmitted through the subject according to the thickness distribution; and a point diffusion function according to the thickness distribution. The second characteristic may include at least one of the following: the energy of the radiation according to the second imaging conditions; a radiation attenuation coefficient according to the thickness distribution of the training subject for an object interposed between the training subject and the radiation detector that detects the radiation transmitted through the training subject when the training radiation image is acquired; the ratio of the scattered radiation component contained in the radiation transmitted through the training subject according to the thickness distribution of the training subject; and a point diffusion function according to the thickness distribution of the training subject.

[0010] In the radiation image processing apparatus according to this disclosure, the processor may display at least one of the two target radiation images and the result of the estimation process on a display.

[0011] The radiation image processing method disclosed herein is a radiation image processing method that performs estimation processing on radiation images by using an estimation model constructed through learning, The computer acquires two target radiation images with different energy distributions obtained by photographing the subject under first shooting conditions, and a first characteristic corresponding to the first shooting conditions. The second characteristic is obtained according to the second imaging conditions when acquiring the training radiation image for which the estimation model was constructed. Based on at least one of the two target radiation images, the thickness distribution of the subject is derived. The scattered radiation component, which is contained in the radiation that has passed through the subject and is scattered by the subject, is removed from the two target radiation images based on the first characteristic. By weighting and subtracting the two target radiation images from which the scattered radiation components have been removed, a first characteristic soft tissue image representing the soft tissue of the subject and a first characteristic bone tissue image representing the bone tissue of the subject are derived. Based on the first characteristic, the second characteristic, and the body thickness distribution, the first characteristic soft tissue image and the first characteristic bone image are converted into a second characteristic soft tissue image and a second characteristic bone image having contrast based on the second characteristic, By adding the second characteristic soft tissue image and the second characteristic bone image, a processed target radiographic image having contrast based on the second characteristic is derived. The processed target radiation image is input to the estimation model and the estimation process is performed.

[0012] The radiation image processing program disclosed herein is a radiation image processing program that causes a computer to perform estimation processing on radiation images by using an estimation model constructed through learning, A procedure for obtaining two target radiation images with different energy distributions, obtained by photographing a subject under first shooting conditions, and a first characteristic corresponding to the first shooting conditions, A procedure for obtaining a second characteristic according to the second imaging conditions when acquiring the training radiation image on which the estimation model described above was constructed, A procedure for deriving the body thickness distribution of the subject based on at least one of the two target radiation images, A procedure for removing a scattered ray component scattered by the subject and included in the radiation that has passed through the subject from the two target radiation images based on the first characteristic; A procedure for deriving a first characteristic soft tissue image representing the soft tissue of the subject and a first characteristic bone tissue image representing the bone tissue of the subject by weighted subtraction of the two target radiation images from which the scattered ray component has been removed; A procedure for converting each of the first characteristic soft tissue image and the first characteristic bone tissue image into a second characteristic soft tissue image and a second characteristic bone tissue image having a contrast based on the second characteristic based on the first characteristic, the second characteristic, and the body thickness distribution; A procedure for deriving a processed target radiation image having a contrast based on the second characteristic by adding the second characteristic soft tissue image and the second characteristic bone tissue image; Causing a computer to execute a procedure of inputting the processed target radiation image into the estimation model and performing the estimation process.

[0013] The technology of the present disclosure may be applied to a program product.

Advantages of the Invention

[0014] According to the present disclosure, the estimation process using the estimation model can be performed with high accuracy.

Brief Description of the Drawings

[0015] [Figure 1] A schematic block diagram showing the configuration of a radiation image imaging system to which a radiation image processing apparatus according to an embodiment of the present disclosure is applied [Figure 2] A diagram showing a schematic configuration of a radiation image processing apparatus according to an embodiment of the present disclosure [Figure 3] A diagram showing a functional configuration of a radiation image processing apparatus according to an embodiment of the present disclosure [Figure 4] A diagram for explaining the imaging of a reference object [Figure 5] A diagram showing the spectrum of radiation [Figure 6]A diagram showing the radiation attenuation coefficients of human soft tissues, bone tissues, and aluminum in relation to radiation energy. [Figure 7] A diagram showing the relationship between the thickness of a reference object and the radiation attenuation coefficient. [Figure 8] Diagram showing the scattering model [Figure 9] A schematic diagram showing the processes performed by the conversion and derivation sections. [Figure 10] A diagram showing the display screen for the estimation results. [Figure 11] Flowchart showing the process performed in this embodiment [Modes for carrying out the invention]

[0016] Embodiments of this disclosure will be described below with reference to the drawings. Figure 1 is a schematic block diagram showing the configuration of a radiographic imaging system to which a radiographic imaging processing device according to an embodiment of this disclosure is applied. As shown in Figure 1, the radiographic imaging system according to this embodiment comprises an imaging device 1, an image storage system 9, and a radiographic imaging processing device 10 according to this embodiment. The imaging device 1, the image storage system 9, and the radiographic imaging processing device 10 are connected to the image storage system 9 via a network (not shown).

[0017] The imaging device 1 is an imaging device for performing energy subtraction using the so-called one-shot method, in which radiation such as X-rays emitted from the radiation source 2 and transmitted through the subject H lying supine on the imaging table 3 is irradiated to the first radiation detector 5 and the second radiation detector 6 with varying energies. During imaging, as shown in Figure 1, the scattered radiation removal grid (hereinafter simply referred to as the grid) 4, the first radiation detector 5, the radiation energy conversion filter 7 made of a copper plate or the like, and the second radiation detector 6 are arranged in order from the side closest to the radiation source 2, and the radiation source 2 is driven. The first and second radiation detectors 5 and 6 and the radiation energy conversion filter 7 are in close contact. The grid 4, the first radiation detector 5, the radiation energy conversion filter 7, and the second radiation detector 6 are detachably mounted below the top plate 3A of the imaging table 3 by mounting parts 3B.

[0018] 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. The second radiation detector 6 acquires a second radiation image G2 of the subject H using high-energy radiation, with soft rays removed. The first and second radiation images G1 and G2 are input to the radiation image processing device 10. The first and second radiation images G1 and G2 are examples of two target radiation images according to this disclosure.

[0019] The first and second radiation detectors 5 and 6 are capable of repeatedly recording and reading radiation images. They may be so-called direct-type radiation detectors that generate an electric charge by directly receiving radiation, or they may be so-called indirect-type radiation detectors that first convert radiation into visible light and then convert that visible light into an electric charge signal. Furthermore, as a method for reading out the radiation image signal, it is desirable to use a so-called TFT (thin film transistor) readout method, in which the radiation image signal is read out by turning a TFT switch on and off, or a so-called optical readout method, in which the radiation image signal is read out by irradiating it with reading light. However, other methods may also be used.

[0020] In addition, in the imaging device 1, there are cases where only one radiation detector is attached to the mounting part 3B to image the subject H.

[0021] Grid 4 is constructed by alternating layers of materials that do not transmit radiation, such as lead, with interspace materials that transmit radiation easily, such as aluminum and fiber, at a fine grid density of approximately 4.0 lines / mm. By using Grid 4, the scattered radiation component of the radiation that has passed through subject H can be removed, but it cannot be completely removed. Therefore, the first and second radiation images G1 and G2 include not only the primary radiation component but also the scattered radiation component of the radiation that has passed through subject H.

[0022] The primary component is the signal component of the pixel value represented by the radiation that passes through the object H and reaches the radiation detector without being scattered by the object H. On the other hand, the scattered component is the signal component of the pixel value represented by the radiation that passes through the object H and is scattered by the object H before reaching the radiation detector.

[0023] The image storage system 9 is a system that stores image data of radiographic images acquired by the imaging device 1. For example, the image storage system 9 stores multiple radiographic images of multiple patients. These multiple radiographic images include radiographic images used as training radiographic images when constructing the estimation model described later. The image storage system 9 retrieves images from the stored radiographic images in response to requests from the radiographic image processing device 10 and transmits them to the requesting device. A specific example of the image storage system 9 is a PACS (Picture Archiving and Communication System).

[0024] Next, a radiation image processing apparatus according to this embodiment will be described. First, the hardware configuration of the radiation image processing apparatus according to this embodiment will be described with reference to Figure 2. As shown in Figure 2, the radiation image processing apparatus 10 is a computer such as a workstation, server computer, or personal computer, and includes a CPU (Central Processing Unit) 11, non-volatile storage 13, and memory 16 as a temporary storage area. The radiation image processing apparatus 10 also includes a display 14 such as a liquid crystal display, input devices 15 such as a keyboard and mouse, and a network I / F (Interface) 17 connected to a network not shown. The CPU 11, storage 13, display 14, input devices 15, memory 16, and network I / F 17 are connected to a bus 18. Note that the CPU 11 is an example of a processor in this disclosure.

[0025] The storage 13 is implemented using an HDD (Hard Disk Drive), an SSD (Solid State Drive), and 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 the radiation image processing program 12 from the storage 13, expands it into memory 16, and executes the expanded radiation image processing program 12.

[0026] The radiation image processing program 12 is stored in a memory device of a server computer connected to the network, or in network storage, in a state that allows external access, and is downloaded and installed on the computers comprising the radiation image processing device 10 upon request. Alternatively, it is recorded on a recording medium such as a DVD (Digital Versatile Disc) or CD-ROM (Compact Disc Read Only Memory) and distributed, and then installed from that recording medium on the computers comprising the radiation image processing device 10.

[0027] Next, the functional configuration of the radiation image processing apparatus according to this embodiment will be described. Figure 3 is a diagram showing the functional configuration of the radiation image processing apparatus according to this embodiment. As shown in Figure 3, the radiation image processing apparatus 10 includes an image acquisition unit 21, a scattered radiation removal unit 22, a subtraction unit 23, a conversion unit 24, a derivation unit 25, a display control unit 26, a characteristic derivation unit 27, and an estimation unit 28. The CPU 11 then executes the radiation image processing program 12 and functions as the image acquisition unit 21, the scattered radiation removal unit 22, the subtraction unit 23, the conversion unit 24, the derivation unit 25, the display control unit 26, the characteristic derivation unit 27, and the estimation unit 28.

[0028] The image acquisition unit 21 causes the imaging device 1 to perform energy subtraction imaging of the subject H, thereby acquiring the first radiation image G1 and the second radiation image G2 of the subject H from the first and second radiation detectors 5 and 6. When acquiring the first radiation image G1 and the second radiation image G2, imaging conditions such as the imaging dose, beam quality, tube voltage (kV), SID (Source Image receptor Distance), which is the distance between the radiation source 2 and the surfaces of the first and second radiation detectors 5 and 6, SOD (Source Object Distance), which is the distance between the radiation source 2 and the surface of the subject H, and the presence or absence of a scatter removal grid are set. The imaging conditions can be set by input from the input device 15 by the operator.

[0029] SOD and SID are used to calculate body thickness distribution, as described later. For SOD, it is preferable to acquire it using a TOF (Time Of Flight) camera, for example. For SID, it is preferable to acquire it using a potentiometer, ultrasonic rangefinder, or laser rangefinder, for example.

[0030] The shooting conditions can be set by the operator via input device 15. The set shooting conditions are stored in storage 13. The first and second radiographic images G1 and G2 acquired by the shooting device 1, along with the shooting conditions, are transmitted to and stored in the image storage system 9.

[0031] The scattered radiation removal unit 22 removes scattered radiation components from the first radiation image G1 and the second radiation image G2 acquired by the image acquisition unit 21. The removal of scattered radiation components will be described below. Any method can be used to remove scattered radiation components, such as the method described in Japanese Patent Application Publication No. 2015-043959. The scattered radiation removal process using the method described in Japanese Patent Application Publication No. 2015-043959 will be described below. In the following description, the first and second radiation images from which scattered radiation components have been removed will also be referred to as G1 and G2, respectively.

[0032] First, the scatter removal unit 22 acquires a virtual model of the subject H having an initial thickness distribution Ts(x,y). The virtual model is data that virtually represents the subject H, in which the thickness according to the initial thickness distribution Ts(x,y) is associated with the coordinate position of each pixel in the first radiation image G1. The virtual model of the subject H having an initial thickness distribution Ts(x,y) is assumed to be stored in the storage 13 in advance, but it may also be acquired from an external server where the virtual model is stored.

[0033] Next, the scattered radiation removal unit 22 derives an estimated primary radiation image Ip(x,y) based on the virtual model, which is an estimated primary radiation image obtained by imaging the virtual model, and an estimated scattered radiation image Is(x,y) based on the scattered radiation image obtained by imaging the virtual model, as shown in equations (1) and (2) below. Furthermore, as shown in equation (3) below, the scattered radiation removal unit 22 derives an estimated image Im(x,y) by combining the estimated primary radiation image Ip(x,y) and the estimated scattered radiation image Is(x,y) as an estimated image obtained by imaging the first radiation image G1 obtained by imaging the subject H. Ip(x,y) = Io(x,y)×exp(-μ1Soft(T(x,y))×T(x,y)) (1) Is(x,y) = Io(x,y)×STPR1(T(x,y))*PSF1(T(x,y)) (2) Im(x,y) = Is(x,y)+Ip(x,y) (3)

[0034] Here, (x,y) is the coordinate of the pixel position of the first radiation image G1, Io(x,y) is the pixel value of the first radiation image G1 at pixel position (x,y), Ip(x,y) is the primary component at pixel position (x,y), and Is(x,y) is the scattered component at pixel position (x,y). Note that when deriving the first estimated image Im(x,y), the initial body thickness distribution Ts(x,y) is used as the body thickness distribution T(x,y) in equations (1) and (2).

[0035] Furthermore, μ1Soft(T(x,y)) in equation (1) is an attenuation coefficient corresponding to the thickness distribution (x,y) of the soft tissue of the human body at the pixel position (x,y). μ1Soft(T(x,y)) can be determined in advance experimentally or by simulation and stored in storage 13. Also, STPR1(T(x,y)) in equation (2) is the scatter-to-primary ratio of the scattered dose to the primary dose contained in the radiation after it passes through an object H with a thickness distribution T(x,y). STPR1(T(x,y)) can also be determined in advance experimentally or by simulation and stored in storage 13.

[0036] Furthermore, PSF1(T(x,y)) in equation (2) is a point spread function that represents the distribution of scattered rays spreading from a single pixel according to the body thickness distribution T(x,y), and is defined according to the energy characteristics of the radiation. Also, * is an operator that indicates a convolution operation. PSF1 also changes depending on the distribution of the irradiation field in the imaging device 1, the composition distribution of the subject H, the irradiation dose during imaging, the tube voltage, the imaging distance, and the characteristics of the radiation detectors 5 and 6. For this reason, PSF1 should be experimentally determined in advance for each energy characteristic of the radiation used by the imaging device 1, according to the irradiation field information, subject information, and imaging conditions, and stored in storage 13.

[0037] The attenuation coefficients μ1Soft, STPR1, and PSF1 are examples of first characteristics corresponding to the first imaging conditions in this disclosure.

[0038] Next, the scattered radiation removal unit 22 modifies the initial body thickness distribution Ts(x,y) of the virtual model so that the difference between the estimated image Im and the first radiation image G1 is small. The scattered radiation removal unit 22 updates the body thickness distribution T(x,y), scattered radiation component Is(x,y), and primary radiation component Ip(x,y) by repeatedly deriving the body thickness distribution T(x,y), scattered radiation component Is(x,y), and primary radiation component Ip(x,y) until the difference between the estimated image Im and the first radiation image G1 satisfies a predetermined termination condition. When the termination condition is met, the scattered radiation removal unit 22 subtracts the scattered radiation component Is(x,y) derived by equation (2) from the first radiation image G1. This removes the scattered radiation component contained in the first radiation image G1. The body thickness distribution T(x,y) derived when the termination condition is met is used in various calculations described later.

[0039] Meanwhile, the scattered radiation removal unit 22 performs scattered radiation removal processing on the second radiation image G2 in the same manner as on the first radiation image G1.

[0040] The derivation of the attenuation coefficients μ1Soft and STPR1 is described below. The attenuation coefficients μ1Soft and STPR1 are derived by the characteristic derivation unit 27. In deriving the attenuation coefficients μ1Soft and STPR1, the image acquisition unit 21 acquires a reference image K0 by having the imaging device 1 take a picture of a reference object that simulates a human body. In this case, only one radiation detector is required. If the reference image K0 is stored in the image storage system 9, the image acquisition unit 21 acquires the reference image K0 from the image storage system 9. Also, in the following explanation, the reference numeral "1" is omitted for generalization purposes.

[0041] Figure 4 is a diagram illustrating the imaging of a reference object. As shown in Figure 4, the reference object 35 has sections of varying thickness, such as 5 cm, 10 cm, and 20 cm, and is made of a material with radiotransmittance similar to that of soft tissue (fat and muscle) in the human body. Therefore, the reference object 35 simulates the radiographic properties of the human body. Here, soft tissue is composed of a mixture of muscle and fat in a certain ratio. The mixing ratio of muscle and fat varies depending on gender and body size, but can be defined by the average body fat percentage (25%). Therefore, a material such as acrylic, corresponding to a composition of 0.75 parts muscle to 0.25 parts fat, is used as the reference object 35.

[0042] To acquire the reference image K0, as shown in Figure 4, the reference object 35 is placed on the top plate 3A of the imaging table 3, and the radiation source 2 is driven to irradiate the radiation detector (in this case, the first radiation detector 5) with radiation via the grid 4, thereby the image acquisition unit 21 acquires the reference image K0. The pixel value of each pixel in the reference image K0 includes a primary radiation component based on radiation that traveled in a straight line through the reference object 35 and a scattered radiation component based on radiation scattered by the reference object 35.

[0043] Note that the reference object 35 is not limited to a single object having a different thickness, as shown in Figure 4. Multiple reference objects, each having a different thickness, may be used. In this case, the reference image K0 may be obtained by photographing multiple reference objects at once, or a reference image corresponding to each reference object may be obtained by photographing multiple reference objects separately.

[0044] When acquiring the reference image K0, imaging conditions such as the imaging dose, tube voltage, SID (Source Image receptor Distance), which is the distance between the radiation source 2 and the surfaces of the first and second radiation detectors 5 and 6, and the presence or absence of grid 4 are set.

[0045] The characteristic derivation unit 27 acquires the energy characteristics of the radiation during imaging of the reference object 35 in order to derive the attenuation coefficient μSoft and STPR. The energy characteristics of the radiation may be acquired from the imaging device 1, or the energy characteristics of the radiation may be stored in the image storage system 9 and acquired from the image storage system 9. The nominal value of the energy characteristics of the imaging device 1 may be used, but since there are individual differences in characteristics for each device, it is preferable to measure them in advance using a semiconductor dosimeter.

[0046] Here, the energy characteristics are defined by (i) the spectrum of radiation emitted from radiation source 2, (ii) the relationship between tube voltage [kV] and total filtration amount [mmAl equivalent], and (iii) the relationship between tube voltage [kV] and aluminum half-value layer [mmAl]. The radiation spectrum is a plot of the relationship between radiation energy [keV] and the number of relative radiation photons. Tube voltage represents the maximum value of the generated radiation energy distribution. Total filtration amount is the filtration amount of each component constituting the imaging device 1, such as the radiation generator and collimator in radiation source 2, converted to the thickness of aluminum. A larger total filtration amount indicates a greater effect of beam hardening in imaging device 1 and a greater proportion of high-energy components in the radiation wavelength distribution. The half-value layer is defined by the thickness of aluminum required to attenuate the dose by half with respect to the generated radiation energy distribution. A thicker aluminum half-value layer indicates a greater proportion of high-energy components in the radiation wavelength distribution.

[0047] Figure 5 shows the radiation spectrum. In Figure 5, the spectrum corresponds to a tube voltage of 90 kV and a total filtration amount of 2.5 mmAl. Note that a total filtration amount of 2.5 mmAl corresponds to a half-value layer of 2.96 mmAl.

[0048] The characteristic derivation unit 27 uses the energy characteristics of the radiation to derive the relationship between the thickness of the reference object 35 and the radiation attenuation coefficient of the reference object 35, which reflects the effect of beam hardening of objects present between the reference object 35 and the radiation detector 5.

[0049] The characteristic derivation unit 27 first derives the energy spectrum of the radiation from the acquired energy characteristics using a well-known Tucker approximation formula or the like. If the acquired energy characteristics are the energy spectrum of the radiation, then the acquired energy spectrum can be used as is.

[0050] Furthermore, the characteristic derivation unit 27 uses the radiation attenuation characteristics of soft tissues of the human body to derive a radiation attenuation coefficient that depends on the thickness of the reference object 35 by simulating the radiation spectrum.

[0051] Here, when the energy spectrum of the radiation emitted from radiation source 2 is denoted by Sin(E) and the thickness of reference object 35 is denoted by t, the radiation dose Xbody(t) after passing through reference object 35 can be calculated using the following equation (4) with respect to the radiation attenuation coefficient μSoft(E) of the soft tissues of the human body. Note that the radiation attenuation coefficients of soft tissues, bone tissues, and aluminum of the human body with respect to radiation energy are known, for example, as shown in Figure 6. Aluminum is the interspace material of grid 4. Here, Figure 6 also shows the radiation attenuation coefficient of acrylic (Polymethyl methacrylate, PMMA), which is the material of reference object 35. As shown in Figure 6, the radiation attenuation coefficient of acrylic is approximately the same as that of soft tissues of the human body.

number

[0052] On the other hand, when imaging a reference object 35 with imaging device 1, a top plate 3A and a grid 4 are present between the reference object 35 and the radiation detectors 5 and 6. Assume that the material of the top plate 3A is acrylic and the interspace material of the grid 4 is aluminum. If the radiation attenuation coefficient of acrylic is μPMMA(E) and the thickness of the top plate 3A (i.e., the thickness of the acrylic) is tPMMA, and the radiation attenuation characteristic of aluminum is μAl(E) and the thickness of the grid 4 (i.e., aluminum) is tAl, then the X-ray dose Xout(t) after passing through the top plate 3A and grid 4 is expressed by the following equation (5).

number

[0053] If the material of the top plate 3A and the interspace material of the grid 4 are unknown, the X-ray dose Xout(t) after passing through the top plate 3A and grid 4 cannot be derived using equation (5) above. In this case, the energy characteristics (kV, TF0) of the radiation emitted from the radiation source 2 and the energy characteristics (kV, TF1) of the radiation after passing through the top plate 3A and grid 4 can be measured using a dosimeter, and the X-ray dose Xout(t) after passing through the top plate 3A and grid 4 can be derived using the following equation (5-1) which uses the energy characteristics (kV, TF0) and (kV, TF1). Note that the energy characteristics in equation (5-1) represent the total filtration amount [mmAl equivalent] of the radiation emitted at a certain tube voltage [kV].

number

[0054] In the imaging system including the top plate 3A and grid 4, the radiation attenuation coefficient of the reference object 35 is expressed as an exponential decay of the rate attenuation of radiation after passing through the reference object 35, as shown in equation (6) below, with the radiation dose when the reference object 35 is absent (i.e., when the thickness of the reference object 35 is 0) as the reference.

number

[0055] By solving equation (6) for the radiation attenuation coefficient μSoft(t) of soft tissue as shown in equation (7) below, the relationship between the thickness t of the reference object 35 and the radiation attenuation coefficient can be derived.

number

[0056] The reference object 35 has several different thicknesses in stages. Therefore, the characteristic derivation unit 27 derives a radiation attenuation coefficient for each of the multiple thicknesses of the reference object 35 using equation (7). Then, for the radiation attenuation coefficients of thicknesses not present in the reference object 35, the characteristic derivation unit 27 derives the relationship between the thickness t of the reference object 35 and the radiation attenuation coefficient by performing an interpolation calculation using the radiation attenuation coefficients of thicknesses present in the reference object 35. Figure 7 shows the relationship between the thickness t of the reference object 35 and the radiation attenuation coefficient. In Figure 7, the relationship between the thickness of the reference object 35 and the radiation attenuation coefficient is shown for a tube voltage of 90kV and a total filtration amount of 2.5mmAl. The characteristic derivation unit 27 derives the relationship between the thickness of the reference object 35 and the radiation attenuation coefficient for each radiation energy characteristic and stores the derived relationship in the storage 13.

[0057] Furthermore, the characteristic derivation unit 27 derives a radiation attenuation coefficient corresponding to the thickness of the reference object 35 based on the relationship between the derived thickness of the reference object 35 and the radiation attenuation coefficient. In addition, it derives the primary line component contained in the reference image K0 based on the radiation attenuation coefficient corresponding to the thickness of the reference object 35.

[0058] Here, when the pixel value of each pixel in the reference image K0 is I0o(x,y), the thickness of the reference object 35 corresponding to each pixel in the reference image K0 is T0(x,y), and the radiation attenuation coefficient derived by equation (7) above for the thickness T0(x,y) of each pixel in the reference image K0 is μSoft0(x,y), the characteristic derivation unit 27 derives the primary linear component I0p(x,y) included in the pixel value of each pixel in the reference image K0 by equation (8) below. Since the reference object 35 has multiple thicknesses in stages, the characteristic derivation unit 27 derives the primary linear component I0p(x,y) for each thickness in the reference object 35. In addition, for primary linear components corresponding to thicknesses not present in the reference object 35, the characteristic derivation unit 27 may derive the relationship between the thickness of the reference object 35 and the primary linear component by performing an interpolation calculation using the primary linear component of the thickness present in the reference object 35. I0p(x,y) = I0o(x,y)×exp(-μSoft0(x,y)×T0(x,y)) (8)

[0059] Furthermore, the characteristic derivation unit 27 derives the scattered radiation components contained in the reference object 35 based on the difference between the pixel values ​​of the reference image K0 and the primary radiation components. That is, the characteristic derivation unit 27 derives the scattered radiation components I0s(x,y) using the following equation (9). Since the reference object 35 has multiple thicknesses in stages, the scattered radiation components I0s(x,y) corresponding to the stages of thickness of the reference object 35 are derived. For scattered radiation components corresponding to thicknesses not present in the reference object 35, the characteristic derivation unit 27 can derive the relationship between the thickness of the reference object 35 and the scattered radiation components by performing interpolation calculations using the scattered radiation components of the thicknesses present in the reference object 35. I0s(x,y) = I0o(x,y)-I0p(x,y) (9)

[0060] The characteristic derivation unit 27 derives the ratio of the scattered radiation component I0s(x,y) to the primary radiation component I0p(x,y) (i.e., I0s(x,y) / I0p(x,y)) as the STPR for each thickness of the reference object 35. Since the thickness of the reference object 35 varies in stages, the STPR for thicknesses not present in the reference object 35 can be derived by interpolation calculation using the STPR for thicknesses present in the reference object 35.

[0061] Figure 8 shows the relationship between the thickness of a reference object and STPR. In Figure 8, the relationship between the thickness of the reference object 35 and STPR is shown when the tube voltage is 90kV and the total filtration amount is 2.5mmAl. The characteristic derivation unit 27 stores the derived scattered radiation model in the storage 13. The reference object 35 simulates the radiation characteristics of the human body. Therefore, the relationship between the thickness of the reference object and STPR shown in Figure 8 represents the relationship between the thickness of the subject H and STPR.

[0062] The relationship between the thickness of the reference object and the STPR can be derived for each energy characteristic of the radiation that the radiation source 2 of the imaging device 1 can emit, and stored in storage 13.

[0063] In this embodiment, in addition to the radiation attenuation coefficient of soft tissue, the radiation attenuation coefficient of bone tissue is also used. Therefore, the characteristic derivation unit 27 also derives the radiation attenuation coefficient of bone tissue. The derivation of the radiation attenuation coefficient μ1Bone of bone tissue will be described below. Note that the radiation attenuation coefficient μ1Bone of bone tissue is also an example of the first characteristic in this disclosure. Also, in the following description, the reference numeral "1" is omitted for generalization purposes.

[0064] Here, assuming a state where soft tissue and bone tissue overlap in the radiation transmission path, and letting the thickness of the soft tissue be tSoft, the radiation attenuation coefficient can be derived as a function dependent on the thickness of the soft tissue. When the energy spectrum of the radiation emitted from radiation source 2 is Sin(E) and the thickness of the soft tissue of subject H is tSoft, the radiation dose Xout1(tSoft) after passing through subject H, assuming there is no bone tissue, can be calculated for each thickness t of subject H using the following equation (10), with respect to the radiation attenuation coefficient μSoft(E) of the soft tissue of the human body. Note that in equation (10), as in equation (5) above, the radiation attenuation coefficients of objects (i.e., the top plate 3A and grid 4) located between subject H and radiation detectors 5 and 6 are taken into consideration.

number

[0065] The radiation dose Xout2(t) in the case of bone tissue is further derived using the radiation attenuation coefficient μBone(E) for bone tissue by the following equation (11).

number

[0066] The radiation attenuation coefficient for bone tissue is expressed as an exponential attenuation of the radiation dose due to bone tissue, using the radiation dose in the absence of bone tissue as the baseline, as shown in equation (12) below.

number

[0067] By solving equation (12) for μBone(t) as shown in equation (13) below, the relationship between the thickness t of the subject H and the radiation attenuation coefficient of the bone tissue can be derived. Note that tBone is the thickness of the bone tissue.

number

[0068] Furthermore, as mentioned above, the PSF also changes depending on the distribution of the irradiation field in the imaging device 1, the composition distribution of the subject H, the irradiation dose during imaging, the tube voltage, the imaging distance, and the characteristics of the radiation detectors 5 and 6. For this reason, the characteristic derivation unit 27 should experimentally determine the PSF in advance for each energy characteristic of the radiation used by the imaging device 1, according to the irradiation field information, subject information, and imaging conditions, and store it in the storage 13.

[0069] Next, the estimation unit 28 will be described. The estimation unit 28 performs estimation processing, for example, to estimate the region of a specific organ contained in a radiographic image, and to estimate the region of a lesion. For this purpose, the estimation unit 28 has an estimation model 28A constructed by machine learning a neural network. The estimation model 28A is trained by using multiple training radiographic images and masks of regions such as lesions in the training radiographic images as ground truth data.

[0070] In this embodiment, the training radiation image is obtained by photographing the subject under second shooting conditions that are different from the first shooting conditions used to acquire the radiation image to be processed. Therefore, the estimation model 28A is constructed by learning using the training radiation image having second characteristics corresponding to the second shooting conditions.

[0071] In this embodiment, the characteristic derivation unit 27 derives the radiation attenuation coefficient μ2Soft for soft tissue, the radiation attenuation coefficient μ2Bone for bone tissue, STPR2, and PSF2 for the training radiation image, in the same manner as described above, according to the second imaging conditions when the training radiation image was acquired. The derived radiation attenuation coefficient μ2Soft for soft tissue, the radiation attenuation coefficient μ2Bone for bone tissue, STPR2, and PSF2 can be stored in the storage 13. The radiation attenuation coefficient μ2Soft for soft tissue and the radiation attenuation coefficient μ2Bone, STPR2, and PSF2 for bone tissue for the training radiation image are examples of second characteristics according to the second imaging conditions in this disclosure.

[0072] The subtraction unit 23 performs energy subtraction processing to derive a bone image Gb, in which the bone portion of the subject H is extracted, and a soft tissue image Gs, in which the soft tissue portion is extracted, from the first and second radiation images G1 and G2, which have undergone scattered radiation removal processing. The bone image Gb and soft tissue image Gs are examples of the first bone image and first soft tissue image according to this disclosure. In subsequent processing, the first and second radiation images G1 and G2 are radiation images from which the scattered radiation component has been removed.

[0073] In deriving the bone image Gb, the subtraction unit 23 generates a bone image Gb in which the bone portion of subject H contained in each of the first and second radiation images G1 and G2 is extracted by performing weighted subtraction between corresponding pixels on the first and second radiation images G1 and G2, as shown in equation (14) below. In equation (14), α1 is a weighting coefficient, and is set to a value that allows the bone portion of subject H contained in each of the radiation images G1 and G2 to be extracted, based on the radiation attenuation coefficients of bone tissue and soft tissue. Gb(x, y)=G1(x, y)-α1×G2(x, y) (14)

[0074] On the other hand, when deriving a soft tissue image Gs, the subtraction unit 23 generates a soft tissue image Gs in which the soft tissue of the subject H included in each radiation image G1 and G2 is extracted by performing weighted subtraction between corresponding pixels on the first and second radiation images G1 and G2, as shown in equation (15) below. In equation (15), α2 is a weighting coefficient, and is set to a value that allows the extraction of the soft tissue of the subject H included in each radiation image G1 and G2 according to equation (15), based on the radiation attenuation coefficients of the bone tissue and soft tissue. The bone image Gb and soft tissue image Gs are examples of the first characteristic bone image and first characteristic soft tissue image according to this disclosure. Gs(x, y)=G1(x, y)-α2×G2(x, y) (15)

[0075] Next, we will explain the processes performed by the conversion unit 24 and the derivation unit 25. Figure 9 is a schematic diagram showing the processes performed by the conversion unit 24 and the derivation unit 25. First, we will explain the processes performed by the conversion unit 24.

[0076] The bone image Gb and soft tissue image Gs derived by the subtraction unit 23 are derived from the first and second radiographic images G1 and G2 acquired by the imaging device 1 under first imaging conditions, and therefore have contrast based on first characteristics (i.e., radiation energy, radiation attenuation coefficient, STPR1, and PSF1) corresponding to the first imaging conditions. The conversion unit 24 converts the bone image Gb and soft tissue image Gs derived by the subtraction unit 23 so that they have contrast based on second characteristics corresponding to the second imaging conditions when the training radiographic image used to construct the estimation model 28A was acquired.

[0077] The conversion unit 24 converts the contrast of the bone image Gb using the following equation (16) to derive the converted bone image Gbt. The conversion unit 24 also converts the contrast of the soft tissue image Gs using the following equation (17) to derive the converted soft tissue image Gst. In equation (16), β1 is derived from β1 = μ2Soft(T(x,y)) / μ1Soft(T(x,y)), and in equation (17), β2 is derived from β2 = μ2Bone(T(x,y)) / μ1Bone(T(x,y)). The body thickness distribution T(x,y) is the one derived by the scatter removal unit 22. The converted bone image Gbt and converted soft tissue image Gst are examples of the second characteristic bone image and second characteristic soft tissue image according to this disclosure. Gbt(x,y) = β1 × Gb(x,y) (16) Gst(x,y) = β² × Gs(x,y) (17)

[0078] The derivation unit 25 derives a composite radiographic image Gc by adding the converted bone image Gbt and converted soft tissue image Gst derived by the conversion unit 24 to their corresponding pixels.

[0079] Here, since the bone image Gb and soft tissue image Gs are derived from the first and second radiographic images G1 and G2 from which the scattered radiation component has been removed, the converted bone image Gbt and converted soft tissue image Gst, as well as the composite radiographic image Gc, do not contain the scattered radiation component. For this reason, the composite radiographic image Gc may be used directly for comparative interpretation with the first and second radiographic images G1 and G2 or the bone image Gb and soft tissue image Gs, but in this embodiment, the scattered radiation component according to the second characteristic is added to the composite radiographic image Gc.

[0080] For this purpose, the derivation unit 25 uses the second characteristics, namely STPR2 and PSF2, corresponding to the second imaging conditions when acquiring the training radiation image, to derive a scattered radiation image Isc representing the scattered radiation component according to the second characteristics using the following equation (18). The scattered radiation component according to the second characteristics is the scattered radiation component corresponding to the scattered radiation component included in the training radiation image used when constructing the estimation model 28A. In equation (18), Gc(x,y) is the pixel value of each pixel in the composite radiation image Gc. The body thickness distribution T(x,y) is the one derived by the scattered radiation removal unit 22. Isc(x,y) = Gc(x,y)×STPR2(T(x,y))*PSF2(T(x,y)) (18)

[0081] The derivation unit 25 then derives the processed target radiation image Gp by adding the composite radiation image Gc and the scattered radiation image Isc with corresponding pixels.

[0082] The estimation unit 28 inputs the processed target radiographic image Gp into the estimation model 28A to derive estimation results for the first and second radiographic images G1 and G2. The estimation results include, for example, the area of ​​the lesion or the area of ​​the organ contained in the first and second radiographic images G1 and G2.

[0083] The display control unit 26 displays the estimation results on the display 14. Figure 10 shows the display screen of the estimation results. As shown in Figure 10, the display screen 40 has a first display area 41 for displaying a first radiographic image G1 or a second radiographic image G2, and a second display area 42 for displaying the estimation results. For example, the first radiographic image G1 is displayed in the first display area 41. The estimation results estimated by the estimation unit 28 are displayed in the second display area 42. In Figure 10, a mark 43 is displayed as the estimation result in the area of ​​the lesion in the first radiographic image G1.

[0084] Next, the processing performed in this embodiment will be described. Figure 11 is a flowchart showing the processing performed in this embodiment. The first and second radiation images G1 and G2 are assumed to be acquired by the imaging device 1 and stored in the storage 13. The first and second characteristics are assumed to be acquired by the characteristic derivation unit 27 and stored in the storage 13. When an instruction to start processing is input from the input device 15, the image acquisition unit 21 acquires the first and second radiation images G1 and G2 from the storage 13 (radiation image acquisition; step ST1). Next, the scattered radiation removal unit 22 derives the body thickness distribution of the subject H from the first and second radiation images G1 and G2 (step ST2), and removes the scattered radiation component from each of the first and second radiation images G1 and G2 (step ST3).

[0085] Next, the subtraction unit 23 derives a bone image Gb, in which the bone portion of the subject H is extracted, and a soft tissue image Gs, in which the soft tissue portion is extracted, from the first and second radiation images G1 and G2, from which the scattered radiation component has been removed (subtraction; step ST4).

[0086] Next, the conversion unit 24 converts the contrast of the bone image Gb and the soft tissue image Gs to derive the converted bone image Gbt and the converted soft tissue image Gst (conversion; step ST5). Then, the derivation unit 25 derives the composite radiation image Gc by adding the converted bone image Gbt and the converted soft tissue image Gst derived by the conversion unit 24 (step ST6). The derivation unit 25 also derives the scattered radiation image Isc, which represents the scattered radiation component according to the second characteristic (step ST7). Finally, the derivation unit 25 derives the processed target radiation image Gp by adding the composite radiation image Gc and the scattered radiation image Isc (step ST8).

[0087] Next, the estimation unit 28 derives an estimation result for the processed target radiation image Gp using the estimation model 28A (step ST9). Then, the display control unit 26 displays the first radiation image G1 or the second radiation image G2 and the estimation result on the display 14 (estimated result display; step ST10), and the process ends.

[0088] Here, the radiographic image acquired by the imaging device 1 includes scattered radiation components corresponding to the characteristics of the radiation energy at the time the image was acquired, the top plate of the imaging platform on which the subject H is placed in the imaging device 1, and the scattered radiation removal grid for removing scattered radiation components contained in the radiation that passed through the subject H when the image was acquired. Therefore, the radiographic image acquired by the imaging device 1 has contrast based on first characteristics corresponding to first imaging conditions at the time the image was acquired.

[0089] On the other hand, the training radiation image used to construct the estimated model 28A includes scattered radiation components corresponding to the characteristics of the radiation energy at the time the training radiation image was acquired, the top plate of the imaging platform on which the subject H in the imaging device 1 is placed at the time the training radiation image was acquired, and the scattered radiation removal grid for removing scattered radiation components contained in the radiation transmitted through the subject H at the time the training radiation image was acquired. Therefore, the training radiation image has contrast based on a second characteristic corresponding to the second imaging conditions at the time the training radiation image was acquired.

[0090] In this embodiment, the scattered radiation component is removed from the first and second radiation images G1 and G2 acquired by the imaging device 1, and the bone image Gb and soft tissue image Gs are derived from the first and second radiation images G1 and G2 from which the scattered radiation component has been removed. Furthermore, based on the first characteristic, the second characteristic, and the body thickness distribution T(x,y) of the subject H, the bone image Gb and soft tissue image Gs are transformed to have a contrast based on the second characteristic corresponding to the second imaging conditions when the training radiation image was acquired, and the transformed bone image Gbt and transformed soft tissue image Gst are combined to derive a combined radiation image Gc.

[0091] Therefore, in this embodiment, the radiation image acquired by the imaging device 1 can be converted to have a contrast corresponding to the training radiation image on which the estimation model 28A was constructed. This makes it possible to match the contrast of the processed target radiation image Gp input to the estimation model 28A with the contrast of the training radiation image. Consequently, according to this embodiment, the estimation results by the estimation model 28A can be obtained with high accuracy.

[0092] Furthermore, based on the second characteristic and the thickness distribution, a scattered radiation image representing the scattered radiation component corresponding to the second characteristic is derived, and by using the derived scattered radiation image to derive the processed target radiation image Gp, the processed target radiation image Gp, including the scattered radiation component, can be made to match the core of the training radiation image.

[0093] In the above embodiment, the estimation model 28A is constructed using a training radiation image having contrast corresponding to a second characteristic corresponding to the second shooting condition. On the other hand, the processing in this embodiment makes it possible to convert the contrast corresponding to the first characteristic of the radiation image acquired by the shooting device 1 to have contrast corresponding to a second characteristic different from the first characteristic. Therefore, processed target radiation images acquired under various shooting conditions can be derived using the first and second radiation images G1 and G2 acquired in a single shooting. By constructing the estimation model 28A using processed target radiation images acquired under such various shooting conditions, the estimation model 28A can be constructed to derive highly accurate estimation results regardless of the shooting conditions. Therefore, it becomes possible to perform highly accurate estimation processing regardless of the shooting conditions of the target radiation image.

[0094] Furthermore, in each of the above embodiments, the first and second radiation images G1 and G2 are acquired by the one-shot method when performing energy subtraction processing, but the method is not limited to this. The first and second radiation images G1 and G2 may also be acquired by the 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 subject H included in the first radiation image G1 and the second radiation image G2 may shift due to the movement of subject H. For this reason, it is preferable to perform alignment of the subject in the first radiation image G1 and the second radiation image G2 before performing the processing of this embodiment.

[0095] Furthermore, in the above embodiment, bone disease prediction processing is performed using the acquired radiation images in a system that captures first and second radiation images G1 and G2 of a subject H using first and second radiation detectors 5 and 6. However, the technology of this disclosure can also be applied when acquiring the first and second radiation images G1 and G2 using a accumulative phosphor sheet instead of radiation detectors. In this case, two accumulative phosphor sheets are stacked and radiation transmitted through the subject H is irradiated to accumulate and record the radiation image information of the subject H on each accumulative phosphor sheet, and the first and second radiation images G1 and G2 can be acquired by photoelectrically reading the radiation image information from each accumulative phosphor sheet. Note that even when acquiring the first and second radiation images G1 and G2 using a accumulative phosphor sheet, the two-shot method may also be used.

[0096] Furthermore, the radiation in the above embodiment is not particularly limited, and in addition to X-rays, alpha rays or gamma rays can be used.

[0097] In this embodiment, each process is executed on any computer. Furthermore, any computer may execute these processes using a processor as hardware, a program as software, or a combination thereof. In that case, the processor is configured to work in cooperation with the program to execute the various processes in this embodiment, and can function as a unit or means in this embodiment. Also, the execution order of the processes by the processor is not limited to the order described and may be changed as appropriate. Any computer may be a general-purpose computer, a computer designed for a specific purpose, a workstation, or any other system capable of executing each process.

[0098] A processor may consist of one or more hardware components, and the type of hardware is not limited. For example, a processor may consist of a CPU (Central Processing Unit), an MPU (Micro Processing Unit), a programmable logic device such as an FPGA (Field Programmable Gate Array), a dedicated circuit for executing a specific process such as an ASIC (Application Specific Integrated Circuit), a GPU (Graphic Processing Unit), or an NPU (Neural Processing Unit). Furthermore, the type of hardware may be a combination of different types of hardware. When multiple hardware components are configured to execute one or more processes of a given processor, these components may reside in physically separate devices or in the same device. Also, in any embodiment, the order of each process performed by the processor is not limited to the order described above and may be changed as appropriate. Hardware is composed of electrical circuits (circuitry) that combine circuit elements such as semiconductor elements.

[0099] Furthermore, the program may be firmware or software such as microcode. Alternatively, the program may be, for example, a set of program modules, each function of which may be implemented by a processor configured to perform its respective function. The program may be program code or multiple code segments stored on one or more non-temporary computer-readable media (e.g., storage media or other storage). The program may be divided and stored on multiple non-temporary computer-readable media located in physically separate devices. Program code or code segments may represent any combination of procedures, functions, subprograms, routines, subroutines, modules, software packages, classes, or instructions, data structures, or program statements. Program code or code segments may be connected to other code segments or hardware circuits by sending and receiving information, data, arguments, parameters, or memory contents.

[0100] Furthermore, although the above embodiment describes a configuration in which the radiation image processing program 12 is pre-stored (installed) in the storage 13, the invention is not limited to this configuration. The radiation image processing program 12 may be provided in the form of a recording medium such as a CD-ROM (Compact Disc Read Only Memory), DVD-ROM (Digital Versatile Disc Read Only Memory), or USB (Universal Serial Bus) memory. Alternatively, the radiation image processing program 12 may be provided in the form of a download from an external device via a network.

[0101] The technology disclosed herein extends to all program products. Program products include all forms of products for providing programs. For example, program products include programs provided via networks such as the Internet, and non-temporary computer-readable recording media such as CD-ROMs, DVDs, and USB memory sticks on which programs are stored.

[0102] The following are additional notes to this disclosure. (Additional note 1) A radiation image processing device that performs estimation processing on radiation images by using an estimation model constructed through learning, Equipped with a processor, The aforementioned processor, Two target radiation images with different energy distributions are obtained by photographing the subject under the first shooting conditions, and a first characteristic corresponding to the first shooting conditions is obtained. The second characteristic is obtained according to the second imaging conditions when acquiring the training radiation image for which the estimation model was constructed. Based on at least one of the two target radiation images, the thickness distribution of the subject is derived. The scattered radiation component, which is contained in the radiation that has passed through the subject and is scattered by the subject, is removed from the two target radiation images based on the first characteristic. By weighting and subtracting the two target radiation images from which the scattered radiation components have been removed, a first characteristic soft tissue image representing the soft tissue of the subject and a first characteristic bone tissue image representing the bone tissue of the subject are derived. Based on the first characteristic, the second characteristic, and the body thickness distribution, the first characteristic soft tissue image and the first characteristic bone image are converted into a second characteristic soft tissue image and a second characteristic bone image having contrast based on the second characteristic, By adding the second characteristic soft tissue image and the second characteristic bone image, a processed target radiographic image having contrast based on the second characteristic is derived. A radiation image processing device that inputs the processed target radiation image into the estimation model and performs the estimation process. (Additional note 2) The processor derives a scattered radiation component corresponding to the second characteristic based on the second characteristic and the thickness distribution. The radiation image processing apparatus according to Appendix 1, further comprising using the derived scattered radiation components to derive the processed target radiation image. (Additional note 3) The first characteristic includes at least one of the following: the energy of the radiation according to the first shooting conditions; a radiation attenuation coefficient according to the thickness distribution of an object interposed between the subject and the radiation detector that detects the radiation transmitted through the subject when acquiring the two target radiation images; the ratio of the scattered radiation component contained in the radiation transmitted through the subject according to the thickness distribution; and a point diffusion function according to the thickness distribution. The radiation image processing apparatus according to appendix 1 or 2, wherein the second characteristic includes at least one of the following: the energy of the radiation according to the second shooting conditions; a radiation attenuation coefficient according to the training body thickness distribution of the training subject for an object interposed between the training subject and the radiation detector that detects the radiation transmitted through the training subject when the training radiation image is acquired; the ratio of the scattered radiation component contained in the radiation transmitted through the training subject according to the training body thickness distribution; and a point diffusion function according to the training body thickness distribution. (Additional note 4) The radiographic image processing apparatus according to any one of the appendices 1 to 3, wherein the processor displays at least one of the two target radiographic images and the result of the estimation process on a display. (Additional note 5) A radiation image processing method that performs estimation processing on radiation images by using an estimation model constructed through learning, The computer acquires two target radiation images with different energy distributions obtained by photographing the subject under first shooting conditions, and a first characteristic corresponding to the first shooting conditions. The second characteristic is obtained according to the second imaging conditions when acquiring the training radiation image for which the estimation model was constructed. Based on at least one of the two target radiation images, the thickness distribution of the subject is derived. The scattered radiation component, which is contained in the radiation that has passed through the subject and is scattered by the subject, is removed from the two target radiation images based on the first characteristic. By weighting and subtracting the two target radiation images from which the scattered radiation components have been removed, a first characteristic soft tissue image representing the soft tissue of the subject and a first characteristic bone tissue image representing the bone tissue of the subject are derived. Based on the first characteristic, the second characteristic, and the body thickness distribution, the first characteristic soft tissue image and the first characteristic bone image are converted into a second characteristic soft tissue image and a second characteristic bone image having contrast based on the second characteristic, By adding the second characteristic soft tissue image and the second characteristic bone image, a processed target radiographic image having contrast based on the second characteristic is derived. A radiation image processing method that inputs the processed target radiation image into the estimation model and performs the estimation process. (Additional note 6) A radiation image processing program that uses an estimation model built through learning to cause a computer to perform estimation processing on radiation images, A procedure for obtaining two target radiation images with different energy distributions, obtained by photographing a subject under first shooting conditions, and a first characteristic corresponding to the first shooting conditions, A procedure for obtaining a second characteristic according to the second imaging conditions when acquiring the training radiation image on which the estimation model described above was constructed, A procedure for deriving the body thickness distribution of the subject based on at least one of the two target radiation images, A procedure for removing scattered radiation components scattered by the subject, which are contained in the radiation that has passed through the subject, from the two target radiation images based on the first characteristic, A procedure for deriving a first characteristic soft tissue image representing the soft tissue of the subject and a first characteristic bone tissue image representing the bone tissue of the subject by weighting and subtracting the two target radiation images from which the scattered radiation components have been removed, A procedure for converting the first characteristic soft tissue image and the first characteristic bone image, respectively, into a second characteristic soft tissue image and a second characteristic bone image having contrast based on the second characteristic, based on the first characteristic, the second characteristic, and the body thickness distribution, A procedure for deriving a processed target radiographic image having contrast based on the second characteristic by adding the second characteristic soft tissue image and the second characteristic bone image, A radiation image processing program that causes a computer to perform the procedure of inputting the processed target radiation image into the estimation model and performing the estimation process. [Explanation of Symbols]

[0103] 1. Imaging device 2 Radiation source 3. Shooting platform 3A Top plate 3B Mounting part 4. Scatter Removal Grid 5,6 Radiation detectors 7. Radiation energy conversion filter 9. Image storage system 10. Radiation image processing device 11 CPU 12. Radiation Image Processing Program 13 Storage 14 displays 15 Input Devices 16 memory 17 Network Interface 18 bus 21 Image acquisition unit 22 Scattered radiation removal section 23 Subtraction section 24 Conversion Unit 25 Derivation part 26 Display Control Unit 27 Characteristic Derivation Section 28 Estimation part 28A Estimated Model 35 Reference object 40 display screen 41,42 Image display area Gb bone image Gp-processed target radiation images Gs Soft tissue images

Claims

1. A radiation image processing device that performs estimation processing on radiation images by using an estimation model constructed through learning, Equipped with a processor, The aforementioned processor, Two target radiation images with different energy distributions are obtained by photographing the subject under the first shooting conditions, and a first characteristic corresponding to the first shooting conditions is obtained. The second characteristic is obtained according to the second imaging conditions when acquiring the training radiation image for which the estimation model described above was constructed. Based on at least one of the two target radiation images, the thickness distribution of the subject is derived. The scattered radiation component, which is contained in the radiation that has passed through the subject and is scattered by the subject, is removed from the two target radiation images based on the first characteristic. By weighting and subtracting the two target radiation images from which the scattered radiation components have been removed, a first characteristic soft tissue image representing the soft tissue of the subject and a first characteristic bone tissue image representing the bone tissue of the subject are derived. Based on the first characteristic, the second characteristic, and the body thickness distribution, the first characteristic soft tissue image and the first characteristic bone image are converted into a second characteristic soft tissue image and a second characteristic bone image having contrast based on the second characteristic, By adding the second characteristic soft tissue image and the second characteristic bone image, a processed target radiation image having contrast based on the second characteristic is derived. A radiation image processing device that inputs the processed target radiation image into the estimation model and performs the estimation process.

2. The processor derives a scattered radiation component corresponding to the second characteristic based on the second characteristic and the thickness distribution, The radiation image processing apparatus according to claim 1, further comprising using the derived scattered radiation component to derive the processed target radiation image.

3. The first characteristic includes at least one of the following: the energy of the radiation according to the first shooting conditions; a radiation attenuation coefficient according to the thickness distribution of an object interposed between the subject and the radiation detector that detects the radiation transmitted through the subject when acquiring the two target radiation images; the ratio of the scattered radiation component contained in the radiation transmitted through the subject according to the thickness distribution; and a point diffusion function according to the thickness distribution. The radiation image processing apparatus according to claim 1 or 2, wherein the second characteristic includes at least one of the following: the energy of the radiation according to the second shooting conditions; a radiation attenuation coefficient according to the training body thickness distribution of the training subject for an object interposed between the training subject and the radiation detector that detects the radiation transmitted through the training subject when the training radiation image is acquired; the ratio of the scattered radiation component contained in the radiation transmitted through the training subject according to the training body thickness distribution; and a point diffusion function according to the training body thickness distribution.

4. The radiation image processing apparatus according to claim 1, wherein the processor displays at least one of the two target radiation images and the result of the estimation process on a display.

5. A radiation image processing method that performs estimation processing on radiation images by using an estimation model constructed through learning, The computer acquires two target radiation images with different energy distributions obtained by photographing the subject under first shooting conditions, and a first characteristic corresponding to the first shooting conditions. The second characteristic is obtained according to the second imaging conditions when acquiring the training radiation image for which the estimation model described above was constructed. Based on at least one of the two target radiation images, the thickness distribution of the subject is derived. The scattered radiation component, which is contained in the radiation that has passed through the subject and is scattered by the subject, is removed from the two target radiation images based on the first characteristic. By weighting and subtracting the two target radiation images from which the scattered radiation components have been removed, a first characteristic soft tissue image representing the soft tissue of the subject and a first characteristic bone tissue image representing the bone tissue of the subject are derived. Based on the first characteristic, the second characteristic, and the body thickness distribution, the first characteristic soft tissue image and the first characteristic bone image are converted into a second characteristic soft tissue image and a second characteristic bone image having contrast based on the second characteristic, By adding the second characteristic soft tissue image and the second characteristic bone image, a processed target radiation image having contrast based on the second characteristic is derived. A radiation image processing method that inputs the processed target radiation image into the estimation model and performs the estimation process.

6. A radiation image processing program that uses an estimation model built through learning to cause a computer to perform estimation processing on radiation images, A procedure for obtaining two target radiation images with different energy distributions, obtained by photographing a subject under first shooting conditions, and a first characteristic corresponding to the first shooting conditions, A procedure for obtaining a second characteristic according to the second imaging conditions when acquiring the training radiation image on which the estimation model described above was constructed, A procedure for deriving the body thickness distribution of the subject based on at least one of the two target radiation images, A procedure for removing scattered radiation components scattered by the subject, which are contained in the radiation that has passed through the subject, from the two target radiation images based on the first characteristic, A procedure for deriving a first characteristic soft tissue image representing the soft tissue of the subject and a first characteristic bone tissue image representing the bone tissue of the subject by weighting and subtracting the two target radiation images from which the scattered radiation components have been removed, A procedure for converting the first characteristic soft tissue image and the first characteristic bone image, respectively, into a second characteristic soft tissue image and a second characteristic bone image having contrast based on the second characteristic, based on the first characteristic, the second characteristic, and the body thickness distribution, A procedure for deriving a processed target radiographic image having contrast based on the second characteristic by adding the second characteristic soft tissue image and the second characteristic bone image, A radiation image processing program that causes a computer to perform the procedure of inputting the processed target radiation image into the estimation model and performing the estimation process.

Images