Fat mass derivation device, method, and program

Abstract

To provide a fat mass derivation device, method and program allowing easy acquisition of a visceral fat mass distribution.SOLUTION: A fat mass derivation device includes at least one processor. The processor derives a fat mass distribution of a subject from first and second radiation images acquired by imaging the subject with radiations having different energy distributions, and derives a visceral fat mass distribution of the subject based on a shape of the fat mass distribution in a cross section orthogonal to a body axis of the subject.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present disclosure relates to a body fat percentage derivation device, method, and program.

Background Art

[0002] In order to prevent the occurrence of cardiovascular diseases (such as myocardial infarction, angina pectoris, cerebral infarction, and obstructive arteriosclerosis) caused by hypertension, hyperglycemia, dyslipidemia, etc., it is important to manage the body fat percentage. For this reason, the composition of human body fat is derived by energy subtraction processing using two radiation images obtained by irradiating a subject with two types of radiation having different energy distributions (see Patent Document 1). In addition, in bone diseases such as osteoporosis, a method for calculating the body fat percentage using the DXA method (Dual X-ray Absorptiometry), which is one of the typical bone mineral quantification methods used for the diagnosis of bone mineral content, has also been proposed (see Patent Document 2).

[0003] On the other hand, it has been clarified that the above-mentioned cardiovascular diseases are caused by the accumulation of visceral fat. For this reason, a method has been proposed for acquiring the distribution of the subcutaneous fat region and the distribution of the visceral fat region in the abdominal region using CT (Computed Tomography) images (see Patent Document 3).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

Problems to be Solved by the Invention

[0005] As described in Patent Document 3, if a CT image is used, the distribution of visceral fat mass can be obtained. However, CT images require time for imaging and involve a large amount of exposure to the subject. Therefore, it is desired to obtain the distribution of visceral fat mass more easily than when using CT images.

[0006] The present disclosure has been made in view of the above circumstances, and an object thereof is to enable easy acquisition of the distribution of visceral fat mass.

Means for Solving the Problems

[0007] The fat mass derivation device according to the present disclosure includes at least one processor, The processor derives the fat mass distribution of the subject from a first radiation image and a second radiation image obtained by photographing the subject with radiation having different energy distributions, Based on the shape of the fat mass distribution in a cross-section orthogonal to the body axis of the subject, the visceral fat mass distribution of the subject is derived.

[0008] In the fat mass derivation device according to the present disclosure, the processor derives the fat mass distribution of the subject from a first radiation image and a second radiation image obtained by photographing the subject from the front or the back,

[0009] The visceral fat mass distribution may be derived by separating the fat mass distribution into a subcutaneous fat mass distribution and a visceral fat mass distribution based on the minimum point of the fat mass distribution and the symmetry of the fat mass distribution.

[0010] In the fat mass derivation device according to the present disclosure, the processor may derive the interval between at least one end of the fat mass distribution and the maximum point closest to the end as the subcutaneous fat thickness of the subject, and derive the visceral fat mass distribution from the fat mass distribution based on the subcutaneous fat thickness.

[0011] Also, in the fat mass derivation device according to the present disclosure, the processor derives the visceral fat mass distribution from the fat mass distribution using a learned neural network, The learned neural network may be learned using teacher data including a learning fat mass distribution and a correct visceral fat mass distribution in the learning fat mass distribution.

[0012] Further, in the fat mass deriving device according to the present disclosure, the processor may display a visceral fat image representing the visceral fat mass distribution.

[0013] Further, in the fat mass deriving device according to the present disclosure, the processor may acquire a past visceral fat image of the same subject, and may display the visceral fat image and the past visceral fat image in a comparable manner.

[0014] Further, in the fat mass deriving device according to the present disclosure, the processor may acquire a past visceral fat image of the same subject, and may display the amount of change of the visceral fat image from the past visceral fat image.

[0015] Further, in the fat mass deriving device according to the present disclosure, the processor may derive an index value of the visceral fat mass in a predetermined region of the subject based on the visceral fat mass distribution, and may display the index value.

[0016] In this case, the index value may be at least one of the volume and weight of the visceral fat mass.

[0017] Alternatively, the index value may be the visceral fat percentage of the subject.

[0018] Further, in the fat mass deriving device according to the present disclosure, the processor may acquire a past index value which is at least one past index value of the subject, and may graphically display the change in the index value based on the index value and the past index value.

[0019] Further, in the fat mass deriving device according to the present disclosure, the processor may derive a smoothed fat mass distribution.

[0020] The method for deriving fat content according to the present disclosure derives the fat content distribution of a subject from a first radiation image and a second radiation image obtained by photographing the subject with radiation having different energy distributions, and derives the visceral fat content distribution of the subject based on the shape of the fat content distribution in a cross-section orthogonal to the body axis of the subject.

[0021] Note that the method for deriving fat content according to the present disclosure may be provided as a program for causing a computer to execute it.

Advantages of the Invention

[0022] According to the present disclosure, the distribution of visceral fat content can be easily obtained.

Brief Description of the Drawings

[0023]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Mode for Carrying Out the Invention

[0024] 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 radiation imaging system to which a fat amount derivation device according to an embodiment of the present disclosure is applied. As shown in FIG. 1, the radiation imaging system according to the present embodiment includes an imaging device 1 and a fat amount derivation device 10 according to the present embodiment.

[0025] The imaging device 1 is an imaging device that performs energy subtraction by a so-called one-shot method in which the first radiation detector 5 and the second radiation detector 6 are irradiated with radiation such as X-rays emitted from the radiation source 3 and transmitted through the subject H, respectively, after changing the energy. At the time of imaging, as shown in FIG. 1, the first radiation detector 5, a 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 close to the radiation source 3, and the radiation source 3 is driven. Note that the first and second radiation detectors 5 and 6 and the radiation energy conversion filter 7 are in close contact with each other.

[0026] As a result, in the first radiation detector 5, a first radiation image G1 of the subject H is acquired by low-energy radiation including so-called soft X-rays. Also, in the second radiation detector 6, a second radiation image G2 of the subject H is acquired by high-energy radiation excluding soft X-rays. The first and second radiation images G1 and G2 are input to the fat mass derivation device 10. Both the first and second radiation images G1 and G2 are front images including the abdomen of the subject H. Although in FIG. 1 radiation is irradiated from the back side of the subject H, radiation may be irradiated from the front side of the subject H instead.

[0027] The first and second radiation detectors 5 and 6 can repeatedly record and read out radiation images, and may use a so-called direct type radiation detector that directly receives radiation irradiation to generate charges, or may use a so-called indirect type radiation detector that once converts radiation into visible light and then converts the visible light into a charge signal. Also, as a method for reading out a radiation image signal, it is desirable to use a so-called TFT readout method in which a 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 a radiation image signal is read out by irradiating reading light, but it is not limited to this and other methods may be used.

[0028] Next, the fat content derivation device according to the first embodiment will be described. First, with reference to FIG. 2, the hardware configuration of the fat content derivation device according to the first embodiment will be described. As shown in FIG. 2, the fat content derivation device 10 is a computer such as a workstation, a server computer, and 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 fat content derivation 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. Note that the CPU 11 is an example of the processor in the present disclosure.

[0029] The storage 13 is realized by an HDD (Hard Disk Drive), an SSD (Solid State Drive), a flash memory, or the like. The storage 13 as a storage medium stores a fat content derivation program 12 installed in the fat content derivation device 10. The CPU 11 reads the fat content derivation program 12 from the storage 13, expands it in the memory 16, and executes the expanded fat content derivation program 12.

[0030] Note that the fat content derivation program 12 is stored in a storage device of a server computer connected to the network or a network storage in a state accessible from the outside, and is downloaded and installed in the computer constituting the fat content derivation device 10 in response to a request. Alternatively, it is recorded and distributed on a recording medium such as a DVD (Digital Versatile Disc) or a CD-ROM (Compact Disc Read Only Memory), and is installed in the computer constituting the fat content derivation device 10 from the recording medium.

[0031] Next, the functional configuration of the fat mass derivation device according to the first embodiment will be described. FIG. 3 is a diagram showing the functional configuration of the fat mass derivation device according to the first embodiment. As shown in FIG. 3, the fat mass derivation device 10 includes an image acquisition unit 21, a scattered ray removal unit 22, a subtraction unit 23, a visceral fat mass derivation unit 24, an index value derivation unit 25, and a display control unit 26. Then, the CPU 11 functions as the image acquisition unit 21, the scattered ray removal unit 22, the subtraction unit 23, the visceral fat mass derivation unit 24, the index value derivation unit 25, and the display control unit 26 by executing the fat mass derivation program 12.

[0032] The image acquisition unit 21 causes the imaging device 1 to perform energy subtraction imaging of the subject H, and acquires a first radiation image G1 and a second radiation image G2, which are front images 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, the radiation quality, the tube voltage, the 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, the SOD (Source Object Distance), which is the distance between the radiation source 3 and the surface of the subject H, and the presence or absence of a scattered ray removal grid are set.

[0033] The SOD and SID are used for calculating the body thickness distribution, as will be described later. For the SOD, it is preferably acquired by, for example, a TOF (Time Of Flight) camera. For the SID, it is preferably acquired by, for example, a potentiometer, an ultrasonic distance meter, a laser distance meter, or the like.

[0034] The imaging conditions may be set by input from the input device 15 by the operator.

[0035] Here, each of the first radiographic image G1 and the second radiographic image G2 includes a scattered ray component based on the radiation scattered within the subject H, in addition to the primary ray component of the radiation that has passed through the subject H. Therefore, the scattered ray removal unit 22 removes the scattered ray component from the first radiographic image G1 and the second radiographic image G2. For example, the scattered ray removal unit 22 may apply the method described in Japanese Patent Application Laid-Open No. 2015-043959 to remove the scattered ray component from the first radiographic image G1 and the second radiographic image G2. When using the method described in Japanese Patent Application Laid-Open No. 2015-043959 or the like, the derivation of the body thickness distribution of the subject H and the derivation of the scattered ray component for removing the scattered ray component are performed simultaneously.

[0036] Hereinafter, the removal of the scattered ray component from the first radiographic image G1 will be described, but the removal of the scattered ray component from the second radiographic image G2 can be performed in the same manner. First, the scattered ray removal unit 22 acquires a virtual model of the subject H having the initial body thickness distribution T0(x, y). The virtual model is data that virtually represents the subject H in which the thickness according to the initial body thickness distribution T0(x, y) is associated with the coordinate position of each pixel of the first radiographic image G1. Note that the virtual model of the subject H having the initial body thickness distribution T0(x, y) may be stored in advance in the storage 13 of the fat amount derivation device 10. Further, the scattered ray removal unit 22 may calculate the body thickness distribution T(x, y) of the subject H based on the SID and the SOD included in the imaging conditions. In this case, the initial body thickness distribution T0(x, y) can be obtained by subtracting the SOD from the SID.

[0037] Next, based on the virtual model, the scattered ray removal unit 22 generates, as an estimated image of the first radiographic image G1 obtained by imaging the subject H, an image obtained by synthesizing an estimated primary ray image estimated by imaging the virtual model and an estimated scattered ray image estimated by imaging the virtual model.

[0038] Next, the scattered ray removal unit 22 corrects the initial thickness distribution T0(x, y) of the virtual model so that the difference between the estimated image and the first radiation image G1 becomes small. The scattered ray removal unit 22 repeatedly generates the estimated image and corrects the thickness distribution until the difference between the estimated image and the first radiation image G1 satisfies a predetermined end condition. The scattered ray removal unit 22 derives the thickness distribution at the time when the end condition is satisfied as the thickness distribution T(x, y) of the subject H. Further, the scattered ray removal unit 22 removes the scattered ray component included in the first radiation image G1 by subtracting the scattered ray component at the time when the end condition is satisfied from the first radiation image G1.

[0039] The subtraction unit 23 performs energy subtraction processing to derive a soft tissue image Gs in which the soft part of the subject H is extracted from the first and second radiation images G1 and G2. Note that the first and second radiation images G1 and G2 in the subsequent processing have the scattered ray components removed. When deriving the soft tissue image Gs, the subtraction unit 23 performs weighted subtraction between the corresponding pixels of the first and second radiation images G1 and G2 as shown in the following formula (1) to generate a soft tissue image Gs in which the soft part of the subject H included in each radiation image G1 and G2 is extracted, as shown in FIG. 4. In formula (1), α is a weighting coefficient. In the following description, the left-right direction of the paper surface of the soft tissue image Gs shown in FIG. 4 is the x direction, and the up-down direction is the y direction. Gs(x, y) = G1(x, y) - α × G2(x, y) (1)

[0040] The visceral fat amount derivation unit 24 derives the visceral fat amount. For this purpose, the visceral fat amount derivation unit 24 derives the fat amount distribution in the soft tissue image Gs by deriving the fat amount for each pixel of the soft tissue image Gs. Here, the soft tissue in the human body includes muscle tissue, adipose tissue, blood, and moisture. In the visceral fat amount derivation unit 24 of the first embodiment, the tissues other than the adipose tissue in the soft tissue are regarded as muscle tissue. That is, in the visceral fat amount derivation unit 24 of the first embodiment, the non-fat tissue including blood and moisture is treated as muscle tissue.

[0041] The visceral fat amount derivation unit 24 separates muscle and fat in the soft tissue image Gs by utilizing the difference in the energy characteristics between muscle tissue and fat tissue. Here, the dose of the radiation after passing through the subject H is lower than that of the radiation before incidence on the subject H, which is a human body. Also, since muscle tissue and fat tissue absorb different amounts of energy and have different attenuation coefficients, the energy spectra of the radiation after passing through muscle tissue and the radiation after passing through fat tissue among the radiation after passing through the subject H are different. As shown in FIG. 5, the energy spectra of the radiation passing through the subject H and irradiating each of the first radiation detector 5 and the second radiation detector 6 depend on the body composition of the subject H, specifically, the ratio of muscle tissue to fat tissue. Since fat tissue is more permeable to radiation than muscle tissue, the higher the ratio of muscle tissue compared to fat tissue, the lower the dose of the radiation after passing through the human body.

[0042] Therefore, the visceral fat amount derivation unit 24 separates muscle and fat from the soft tissue image Gs by utilizing the difference in the energy characteristics of the above-described muscle tissue and fat tissue. Then, the visceral fat amount derivation unit 24 generates a fat image from the soft tissue image Gs and derives the fat amount of each pixel based on the pixel value of the fat image.

[0043] Note that the specific method by which the visceral fat amount derivation unit 24 separates muscle and fat from the soft tissue image Gs is not limited. As an example, the visceral fat amount derivation unit 24 of the first embodiment generates a fat image Gf from the soft tissue image Gs according to the following formulas (2) and (3). Specifically, first, the visceral fat amount derivation unit 24 derives the fat ratio rf(x, y) at each pixel position (x, y) in the soft tissue image Gs according to formula (2). Note that μm in formula (2) is a weight coefficient corresponding to the attenuation coefficient of muscle tissue, and μf is a weight coefficient corresponding to the attenuation coefficient of fat tissue. Also, Δ(x, y) represents the concentration difference distribution. The concentration difference distribution is the distribution on the image of the concentration change as seen from the concentration obtained when radiation reaches the first radiation detector 5 and the second radiation detector 6 without passing through the subject H. The distribution of the concentration change on the image is calculated by subtracting the concentration of each pixel in the region of the subject H from the concentration in the void region obtained by directly irradiating the first radiation detector 5 and the second radiation detector 6 with radiation in the soft tissue image Gs. rf(x,y)={μm-Δ(x,y) / T(x,y)} / (μm-μf) (2)

[0044] Furthermore, the visceral fat amount derivation unit 24 generates a fat image Gf from the soft tissue image Gs according to the following formula (3). Note that (x, y) in formula (3) is the pixel position of the fat image Gf and corresponds to the pixel position of the soft tissue image Gs. Gf(x,y)=rf(x,y)×Gs(x,y) (3)

[0045] In this embodiment, the visceral fat amount derivation unit 24 derives the pixel value Gf(x, y) of the fat image Gf as the fat amount. In the fat image Gf, the higher the fat amount, the higher the brightness. The pixel value Gf(x, y) of the fat image Gf is information representing the relative difference in fat at each position of the subject H. Note that since the pixel value Gf(x, y) does not represent the true fat amount, the visceral fat amount derivation unit 24 multiplies each pixel value Gf(x, y) of the fat image Gf by a coefficient K1(x, y) representing the relationship between the predetermined pixel value and the fat amount, as shown in the following formula (4), to obtain the true fat amount F(x, y) (g / cm2 ) may be derived. Also, the fat percentage rf(x, y) for each pixel may be used as the fat mass. In the following description, the pixel value Gf(x, y) is used as the fat mass. F(x, y) = K1(x, y) × Gf(x, y) (4)

[0046] The visceral fat mass derivation unit 24 derives the distribution of the fat mass in the fat image Gf. For this purpose, the visceral fat mass derivation unit 24 derives the distribution of the fat mass in each line in the x-axis direction in the fat image Gf. The distribution of the fat mass in the line in the x-axis direction in the fat image Gf represents the distribution of the fat mass in a cross-section perpendicular to the body axis of the subject H.

[0047] FIG. 6 is a diagram for explaining the distribution of the fat mass in a line extending in the x-axis direction in the fat image Gf. For the sake of explanation, FIG. 6 shows a tomographic image D0 in a cross-section orthogonal to the body axis of the subject H corresponding to the line L0 in the x-axis direction set in the fat image Gf. The fat mass distribution 30 shown in FIG. 6 has the horizontal axis indicating the pixel position of the fat image Gf on the line L0, and the vertical axis indicating the fat mass, that is, the pixel value Gf(x, y) of the fat image Gf.

[0048] Since the fat contained in the subject H is distinguished into subcutaneous fat and visceral fat, the tomographic image D0 shown in FIG. 6 includes a subcutaneous fat region 31 and a visceral fat region 32. In the tomographic image D0 shown in FIG. 6, the hatching directions are different between the subcutaneous fat region 31 and the visceral fat region 32. Here, the value obtained by integrating the pixel values in the subcutaneous fat region 31 and the pixel values in the visceral fat region 32 of the tomographic image D0 in the direction of the arrow A, which is the front-rear direction of the subject H shown in FIG. 6, corresponds to the fat mass distribution 30. When only the pixel values of the subcutaneous fat region 31 in the tomographic image D0 are integrated in the direction of the arrow A, in the fat mass distribution, the fat mass is distributed such that there are maximum values near the left and right ends of the subject H, and the value decreases toward the vicinity of the center of the subject H. In FIG. 6, for the sake of explanation, the fat mass distribution corresponding to the subcutaneous fat in the fat mass distribution 30 is shown by a broken line 33, and the hatching directions are different between the subcutaneous fat distribution and the visceral fat distribution.

[0049] The visceral fat amount derivation unit 24 derives the visceral fat amount based on the shape of the fat amount distribution 30. For this purpose, the visceral fat amount derivation unit 24 first derives the subcutaneous fat amount. FIG. 7 is a diagram for explaining the derivation of the subcutaneous fat amount. The visceral fat amount derivation unit 24 derives the subcutaneous fat amount by utilizing the fact that the subcutaneous fat is symmetrically distributed in the left and right directions in the fat amount distribution 30. For this purpose, the visceral fat amount derivation unit 24 first detects the end points E1 and E2 of the fat amount distribution 30 as shown in FIG. 7. Next, the visceral fat amount derivation unit 24 detects the maximum points Max1 and Max2 that first appear when tracing the fat amount distribution 30 in the inward direction from the end points E1 and E2.

[0050] Subsequently, the visceral fat amount derivation unit 24 detects the minimum points of the fat amount distribution 30 that are between the maximum points Max1 and Max2. In FIG. 7, the minimum points are indicated by black circles. Note that only some of the minimum points are shown in FIG. 7 for the purpose of explanation. Next, the visceral fat amount derivation unit 24 sets a center line C0 that bisects the end points E1 and E2 or the maximum points Max1 and Max2. Then, as shown in FIG. 8, the visceral fat amount derivation unit 24 sets corresponding points corresponding to the minimum points at positions symmetric with respect to the center line C0. In FIG. 8, the corresponding points corresponding to the minimum points are indicated by white circles.

[0051] Next, as shown in FIG. 9, the visceral fat amount derivation unit 24 connects the end points E1 and E2, the maximum points Max1 and Max2, the minimum points, and the corresponding points with line segments. Note that the visceral fat amount derivation unit 24 connects the minimum points and the corresponding points that are close to each other in the axial direction with line segments. And when the two line segments intersect the minimum points and the corresponding points at an upwardly convex angle, the visceral fat amount derivation unit 24 excludes the minimum points and the corresponding points and connects the minimum points and the corresponding points with line segments again. In FIG. 9, the minimum points P1 to P4 and the corresponding points Pt1 to Pt4 are excluded. The visceral fat amount derivation unit 24 repeats this process until there are no line segments that intersect at an upwardly convex angle. When connecting the remaining minimum points and the corresponding points with line segments between the maximum points Max1 and Max2, as shown in FIG. 10, the remaining minimum points and the corresponding points between the maximum points Max1 and Max2 are connected so as to be downwardly convex.

[0052] As shown in FIG. 11, the visceral fat amount derivation unit 24 approximates the minimum points and corresponding points remaining between the maximum point Max1 and the maximum point Max2 with a curve 35. For the curve approximation, for example, a method of approximating a quadratic function by the least squares method can be used, but it is not limited thereto. Then, the visceral fat amount derivation unit 24 derives the fat amount distribution 30 above the curve 35 in the fat amount distribution 30 as the visceral fat amount distribution 36. The visceral fat amount distribution 36 is a distribution of pixel values Gv(x, y) representing the visceral fat amount.

[0053] Note that the visceral fat amount derivation unit 24 derives the two-dimensional visceral fat amount distribution in a predetermined region of the fat image Gf as the visceral fat image Gv by deriving the visceral fat amount distribution 36 in all lines in the x-axis direction in a predetermined region of the fat image Gf. The predetermined region may be the entire region of the fat image Gf or a region preset in the fat image Gf. As the preset region, for example, a region within a range of ±5 cm from the center line that bisects the fat image Gf horizontally may be used, but it is not limited thereto. Also, the fat image Gf may be displayed, and the region specified by the user in the displayed fat image Gf may be used as the predetermined region.

[0054] Also, the predetermined region may be a region of a predetermined site detected from the first or second radiographic image G1, G2. In this case, the abdominal region can be used as the predetermined region. The abdominal region can be set by detecting the region from the lower end of the lung field to the hip joint in the radiographic image. In this case, by performing a process of detecting the lung field and a process of detecting the bone region on the radiographic image, the region between the lower end of the lung field and the hip joint can be set as the predetermined region.

[0055] Thereby, in a predetermined region in the radiographic image of the subject H, a visceral fat image Gv that two-dimensionally represents the visceral fat amount distribution is derived. The pixel value Gv(x, y) of each pixel in the visceral fat image Gv represents the visceral fat amount.

[0056] The index value derivation unit 25 derives an index value of the visceral fat amount in a predetermined region of the subject H based on the visceral fat amount distribution represented by the visceral fat image Gv. As the index value, for example, the volume and weight of visceral fat can be used. When deriving the volume of visceral fat, the index value derivation unit 25 derives the area S0 of the visceral fat amount distribution 36 shown in FIG. 11. Then, by multiplying the derived area S0 by the size Y0 in the y-axis direction of one pixel of the fat image Gf, the volume vi of the visceral fat amount distribution 36 in each line L0 in the x-axis direction is derived. Then, the index value derivation unit 25 integrates all the volumes vi derived for all the lines in the x-axis direction in a predetermined region in the fat image Gf to derive the volume V0 of visceral fat. Further, the index value derivation unit 25 derives the weight of visceral fat by multiplying the derived volume V0 by the standard density of fat.

[0057] The display control unit 26 displays the visceral fat image Gv on the display 14. FIG. 12 is a diagram showing a display screen of the visceral fat image. As shown in FIG. 12, the visceral fat image Gv is displayed on the display screen 40. In the visceral fat image Gv, the distribution of the visceral fat amount is shown in different colors according to the visceral fat amount. In FIG. 12, the difference in color is shown by different hatching. Also, on the right side of the visceral fat image Gv, an index value display area 41 is displayed. In the index value display area 41, the volume and weight derived by the index value derivation unit 25 are displayed.

[0058] Next, the processing performed in the first embodiment will be described. FIG. 13 is a flowchart showing the processing 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, and acquires first and second radiation images G1 and G2 (radiation image acquisition; step ST1). Next, the scattered ray removal unit 22 removes the scattered ray components from the first and second radiation images G1 and G2 (step ST2). Further, the subtraction unit 23 derives a soft tissue image Gs in which the soft tissue of the subject H is extracted from the first and second radiation images G1 and G2 from which the scattered ray components have been removed (step ST3).

[0059] Subsequently, the visceral fat amount derivation unit 24 derives a fat image Gf and further a fat amount distribution 30 from the soft tissue image Gs (step ST4), and derives a visceral fat amount distribution 36 from the fat amount distribution 30 to derive a visceral fat image Gv (step ST5). Further, the index value derivation unit derives an index value of the visceral fat amount from the visceral fat image Gv (step ST6). Then, the display control unit 26 displays the visceral fat image Gv representing the visceral fat amount distribution 36 and the index value on the display 14 (step ST7), and ends the processing.

[0060] Thus, in the present embodiment, the fat amount distribution of the subject H is derived from the radiation image by the energy subtraction process, and the visceral fat amount distribution of the subject is derived based on the shape of the fat amount distribution in the cross section orthogonal to the body axis of the subject. Thus, in the present embodiment, since the radiation image is used without using the CT image, the distribution of the visceral fat amount can be easily obtained.

[0061] Next, a second embodiment of the present disclosure will be described. Note that the fat amount derivation device according to the second embodiment differs from the first embodiment in the method of deriving the visceral fat amount distribution 36, and since the configuration of the device is the same as the configuration of the fat amount derivation device 10 according to the first embodiment, a detailed description of the device configuration will be omitted here.

[0062] Here, as can be seen from the tomographic image D0 and the fat mass distribution 30 in FIG. 6 described above, the regions near the left and right ends of the fat mass distribution 30 are subcutaneous fat regions. In the second embodiment, the visceral fat mass derivation unit 24 is different from the first embodiment in that the visceral fat mass distribution is derived based on the subcutaneous fat region in the fat mass distribution 30.

[0063] FIG. 14 is a diagram for explaining the derivation of the visceral fat mass distribution in the second embodiment. As shown in FIG. 14, the visceral fat mass derivation unit 24 detects the maximum points Max1 and Max2 that first appear when tracing the fat mass distribution 30 in the inward direction from the end points E1 and E2 of the fat mass distribution 30. The maximum points Max1 and Max2 correspond to the positions where the subcutaneous fat is thickest in the fat mass distribution 30. Then, the visceral fat mass derivation unit 24 derives the average value of the distance t1 from the end point E1 to the maximum point Max1 and the distance t2 from the end point E2 to the maximum point Max2 in the fat mass distribution 30 as the thickness of the subcutaneous fat of the subject H (hereinafter referred to as the subcutaneous fat thickness t0).

[0064] Next, the visceral fat mass derivation unit 24 detects the minimum points Min1 and Min2 that first appear when tracing the fat mass distribution 30 in the inward direction from the maximum points Max1 and Max2. The minimum points Min1 and Min2 are the points where the region changes from the region with only subcutaneous fat to the region including subcutaneous fat and visceral fat. Further, the visceral fat mass derivation unit 24 derives the visceral fat mass distribution by subtracting the value corresponding to the subcutaneous fat thickness t0 from the fat mass distribution 30 in the region R0 between the minimum points Min1 and Min2.

[0065] Specifically, the visceral fat amount derivation unit 24 derives the pixel value Q0 of the region directly irradiated with radiation in the first radiation image G1 (or the second radiation image G2) from which the scattered ray component has been removed. Assuming that the attenuation coefficient of radiation per unit thickness of fat is μ(t), the pixel value Q1 obtained by the radiation passing through subcutaneous fat with a thickness of t0 can be calculated by Q1 = Q0 / exp(μ(t) × t0). Therefore, the pixel value ΔQ representing the fat amount corresponding to subcutaneous fat with a thickness of t0 can be calculated by ΔQ = Q0 - Q1. For example, when Q0 = 100000, t0 = 2 cm, and μ(t) = 0.2, Q1 = 100000 / exp(0.2 × 2) = 67032, so ΔQ = 100000 - 67032 = 32968 can be obtained. Note that a predetermined value may be used for the attenuation coefficient μ(t) of fat.

[0066] Then, the visceral fat amount derivation unit 24 derives the visceral fat amount distribution approximately by uniformly subtracting the pixel value ΔQ, which is a value corresponding to the subcutaneous fat thickness t0, from the fat amount distribution 30 in the region R0 between the minimum points Min1 and Min2.

[0067] Next, a third embodiment of the present disclosure will be described. Note that the fat amount derivation device according to the third embodiment differs from the first embodiment in the method of deriving the visceral fat amount distribution 36, and since the configuration of the device is the same as that of the fat amount derivation device according to the first embodiment, a detailed description of the device configuration will be omitted here.

[0068] The visceral fat amount derivation unit 24 of the fat amount derivation device according to the third embodiment differs from the first embodiment in that it uses a learned neural network to derive the visceral fat amount distribution from the fat amount distribution. Here, the learned neural network is composed of, for example, a convolutional neural network, and is constructed by performing machine learning on the convolutional neural network using teacher data including a learning fat amount distribution and the correct visceral fat amount distribution in the learning fat amount distribution.

[0069] FIG. 15 is a diagram showing teacher data. As shown in FIG. 15, the teacher data 51 includes a learning fat amount distribution 52 and a correct visceral fat amount distribution 53 representing the visceral fat amount distribution in the learning fat amount distribution 52. The learning fat amount distribution 52 is derived by integrating the signal values of both the subcutaneous fat and visceral fat regions extracted from the tomographic images included in the CT image in the front-back direction of the human body. Note that the front-back direction of the human body corresponds to the imaging direction of the energy subtraction imaging, that is, the direction in which the subject H is irradiated with radiation. The learning fat amount distribution 52 is derived by integrating the signal values of the visceral fat region extracted from the tomographic images included in the CT image in the front-back direction of the human body.

[0070] Then, the learning fat amount distribution 52 is input into the convolutional neural network, and the difference between the visceral fat amount distribution output from the convolutional neural network and the correct visceral fat amount distribution 53 is derived as a loss, and the parameters of the convolutional neural network are corrected so that the loss becomes smaller, thereby learning the convolutional neural network. Then, the learned neural network is constructed by repeating the learning until the loss reaches a predetermined threshold value. At this time, the convolutional neural network will learn the characteristics of the shapes of the fat amount distribution and the visceral fat amount distribution.

[0071] In the third embodiment, the visceral fat amount derivation unit 24 derives the visceral fat amount distribution from the fat amount distribution 30 derived by the visceral fat amount derivation unit 24 using the learned neural network. In this way, it is also possible to derive the visceral fat amount distribution by using the learned neural network.

[0072] In each of the above embodiments, the index value derivation unit 25 derives at least one of the volume and weight of visceral fat in a predetermined region of the fat image Gf as an index value. However, the index value is not limited to this. Information on the weight of the subject H may be acquired, and the visceral fat percentage of the subject H may be derived as an index value by dividing the weight of the visceral fat by the weight of the subject H. In this case, it is preferable to use the weight of the predetermined region in which the visceral fat amount distribution in the fat image Gf is derived as the weight of the subject H. The weight of the predetermined region is derived as follows.

[0073] First, the index value derivation unit 25 derives the volume Vh0 of a predetermined region of the subject H by integrating the thickness distribution T(x, y) derived by the scattered light removal unit 22 in a predetermined region of the fat image Gf. Further, the index value derivation unit 25 derives the area of the fat amount distribution 30 in each line L0 in the x-axis direction in a predetermined region of the fat image Gf, and multiplies the derived area by the size Y0 in the y-axis direction of one pixel of the fat image Gf to derive the volume vai of the fat amount distribution 30 in each line L0 extending in the x-axis direction. Then, the index value derivation unit 25 derives the fat volume Vh1 by integrating all the volumes vai derived in the x-axis direction in a predetermined region of the fat image Gf. Further, the index value derivation unit 25 derives the volume Vh2 of tissues other than fat, such as muscle, blood, and water, of the subject H by subtracting the fat volume Vh1 from the volume Vh0 of the subject H.

[0074] Then, the index value derivation unit 25 derives the weight of the subject H in a predetermined region by calculating Vh1×the standard density of fat + Vh2×the standard density of tissues other than fat. Note that, for example, the density of water may be used as the standard density of tissues other than fat.

[0075] In addition, in each of the above embodiments, the visceral fat image of the subject H and the index value of the visceral fat amount may be stored in an image storage server (not shown). In this case, in the fat amount derivation device according to the present embodiment, the image acquisition unit 21 may acquire a past visceral fat image of the same subject H from the image storage server and display the current visceral fat image and the past visceral fat image in a comparable manner. Further, the image acquisition unit 21 may acquire the index value of the past visceral fat amount of the same subject H from the image storage server and graphically display the change in the index value based on the current index value of the visceral fat amount and the past index value of the visceral fat amount. Here, the index value of the past visceral fat amount is an example of the past index value.

[0076] FIG. 16 is a diagram showing a display screen of a current visceral fat image and a past visceral fat image. As shown in FIG. 16, on the display screen 40, the current visceral fat image Gv and the past visceral fat image Gv1 are arranged side by side and displayed in a comparable manner. Note that the current visceral fat image Gv and the past visceral fat image Gv1 may be displayed so as to be switchable by an operation from the input device 15. In this way, when the current visceral fat image Gv and the past visceral fat image Gv1 are displayed, in the index value derivation unit 25, a graph representing the transition of the change in the visceral fat amount may be derived based on the current index value of the visceral fat amount and the past index value of the visceral fat amount, and the derived graph may be displayed. Further, a change rate may be derived as the index value. In FIG. 16, a graph 42 and a change rate 43 derived by the index value derivation unit 25 are displayed.

[0077] In addition, when acquiring past visceral fat images, the difference value between corresponding pixels of the current visceral fat image Gv and the past visceral fat image Gv1 may be derived as the change amount of the visceral fat amount. In this case, a change amount image representing the change amount may be displayed on the display 14. FIG. 17 is a diagram showing a display screen of the change amount image. As shown in FIG. 17, on the display screen 40, instead of the current visceral fat image Gv and the past visceral fat image Gv1, a change amount image Gd representing the change amount between the current visceral fat image Gv and the past visceral fat image Gv1 is displayed. In addition to the visceral fat image Gv and the past visceral fat image Gv1 shown in FIG. 16, the change amount image Gd may be displayed.

[0078] Also, in each of the above embodiments, when obtaining the maximum and minimum points of the fat amount in the fat amount distribution 30, that is, the pixel values of the fat image Gf, in order to reduce the influence of large fluctuations in pixel values in units of one pixel due to the influence of random noise, the fat amount distribution 30 may be smoothed. For example, for each pixel in the x-axis direction in the fat amount distribution 30, the moving average of the pixel values with surrounding pixels is calculated, and the fat amount distribution 30 is smoothed by setting the moving average as the pixel value of each pixel, and the maximum and minimum points may be obtained. At this time, the pixel width for obtaining the moving average may be, for example, 3 to 13 pixels.

[0079] Also, in each of the above embodiments, when performing the energy subtraction process for deriving the fat image, the first and second radiation images G1, G2 are acquired by the one-shot method, but it is not limited thereto. The first and second radiation images G1, G2 may 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, due to the body movement of the subject H, the position of the subject H included in the first radiation image G1 and the second radiation image G2 may shift. Therefore, in the first radiation image G1 and the second radiation image G2, it is preferable to perform the alignment of the subject and then perform the processing of this embodiment.

[0080] Also, in each of the above embodiments, in a system for photographing a subject H using the first and second radiation detectors 5 and 6, the visceral fat amount distribution is derived using the first and second radiation images obtained. However, instead of the radiation detector, the visceral fat amount distribution may be derived using the first and second radiation images G1 and G2 obtained using a storage phosphor sheet. In this case, two storage 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 stored and recorded in each storage phosphor sheet. The first and second radiation images G1 and G2 may be obtained by photoelectrically reading the radiation image information from each storage phosphor sheet. Note that the two-shot method may also be used when obtaining the first and second radiation images G1 and G2 using the storage phosphor sheet.

[0081] Also, the radiation in each of the above embodiments is not particularly limited, and in addition to X-rays, α-rays or γ-rays can be used.

[0082] Also, in each of the above embodiments, for example, as the hardware structure of a processing unit (Processing Unit) that executes various processes such as an image acquisition unit 21, a scattered ray removal unit 22, a subtraction unit 23, a visceral fat amount derivation unit 24, an index value derivation unit 25, and a display control unit 26, the following various processors (Processor) can be used. In addition to the CPU, which is a general-purpose processor that executes software (program) and functions as various processing units as described above, the above various processors include a programmable logic device (Programmable Logic Device: PLD), which is a processor whose circuit configuration can be changed after manufacturing, 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).

[0083] One processing unit may be composed of one of these various processors, or may be composed of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs or a combination of a CPU and an FPGA). Also, a plurality of processing units may be configured with one processor.

[0084] As an example of configuring a plurality of processing units with one processor, first, as represented by computers such as clients and servers, one processor is configured by a combination of one or more CPUs and software, and this processor functions as a plurality of processing units. Second, as represented by a System On Chip (SoC), there is a form in which a processor that realizes the functions of the entire system including a plurality of processing units with one IC (Integrated Circuit) chip is used. Thus, various processing units are configured using one or more of the above various processors as a hardware structure.

[0085] Furthermore, as a hardware structure of these various processors, more specifically, an electrical circuit (Circuitry) combining circuit elements such as semiconductor elements can be used.

Explanation of Signs

[0086] 1 Imaging device 3 Radiation source 5, 6 Radiation detector 7 Radiation energy conversion filter 10 Fat mass derivation device 11 CPU 12 Fat mass derivation program 13 Storage 14 Display 15 Input device 16 Memory 17 Network I / F 18 Bus 21 Image acquisition unit 22 Scattered ray removal unit 23 Subtraction unit 24 Visceral fat amount derivation unit 25 Index value derivation unit 26 Display control unit 30 Fat amount distribution 31 Subcutaneous fat region 32 Visceral fat region 33 Dashed line 35 Curve 36 Visceral fat amount distribution 40 Display screen 41 Index value display region 42 Graph 43 Change amount 51 Teacher data 52 Fat amount distribution for learning 53 Correct visceral fat amount distribution C0 Center line D0 Tomographic image E1, E2 End points of fat amount distribution Gv Visceral fat image Gv1 Past visceral fat image Gd Change amount image H Subject L0 Line Max1, Max2 Maximum points Min1, Min2 Minimum points P1~P4 Minimum points Pt1~Pt4 Corresponding points R0 Region between minimum points t1, t2 Thickness of subcutaneous fat

Claims

1. Comprising at least one processor, wherein the processor, derives the fat mass distribution of the subject from a first radiation image and a second radiation image obtained by photographing the subject from the front or the back with radiation having different energy distributions, and derives the visceral fat mass distribution of the subject by separating the fat mass distribution into a subcutaneous fat mass distribution and a visceral fat mass distribution based on the minimum point of the fat mass distribution in a cross-section orthogonal to the body axis of the subject and the symmetry of the shape of the fat mass distribution. A fat mass deriving device.

2. The fat mass deriving device according to claim 1, wherein the processor displays a visceral fat image representing the visceral fat mass distribution.

3. The fat mass deriving device according to claim 2, wherein the processor acquires a past visceral fat image of the same subject, and displays the visceral fat image and the past visceral fat image in a comparable manner.

4. The fat mass deriving device according to claim 2 or 3, wherein the processor acquires a past visceral fat image of the same subject, and displays the amount of change of the visceral fat image from the past visceral fat image.

5. The fat mass deriving device according to any one of claims 1 to 4, wherein the processor derives an index value of the visceral fat mass in a predetermined region of the subject based on the visceral fat mass distribution, and displays the index value.

6. The fat mass deriving device according to claim 5, wherein the index value is at least one of the volume and weight of the visceral fat mass.

7. The fat mass deriving device according to claim 5 or 6, wherein the index value is the visceral fat percentage of the subject.

8. The fat mass deriving device according to any one of claims 5 to 7, wherein the processor acquires a past index value which is at least one past index value of the subject, and graphically displays the change of the index value based on the index value and the past index value.

9. The fat mass deriving device according to any one of claims 1 to 8, wherein the processor derives the smoothed fat mass distribution.

10. Derives the fat mass distribution of the subject from a first radiation image and a second radiation image obtained by photographing the subject from the front or the back with radiation having different energy distributions, A method for deriving visceral fat mass distribution of a subject by separating the fat mass distribution into subcutaneous fat mass distribution and visceral fat mass distribution based on the minimum point of the fat mass distribution and the symmetry of the shape of the fat mass distribution in a cross-section orthogonal to the body axis of the subject.

11. A procedure for deriving the fat mass distribution of a subject from a first radiation image and a second radiation image obtained by photographing the subject from the front or the back with radiation having different energy distributions, and A fat mass derivation program that causes a computer to execute a procedure for deriving the visceral fat mass distribution of the subject by separating the fat mass distribution into subcutaneous fat mass distribution and visceral fat mass distribution based on the minimum point of the fat mass distribution and the symmetry of the shape of the fat mass distribution in a cross-section orthogonal to the body axis of the subject.

Images