Radiation image processing device, its operating method, and radiation image processing program

Abstract

To provide a radiation image processing apparatus, an operation method thereof and a radiation image processing program which, in set imaging of the same person in a radiation image, shorten the time of total energy subtraction processing, and improve and stabilize the accuracy of obtaining ratios of muscle and fat.SOLUTION: A radiation image processing apparatus 12 of the present invention is configured to: acquire at least two radiation images in which radiation energies different from each other are used, respectively, for the same subject H; calculate a fat percentage Rf occupied in a soft tissue of the subject by subtraction processing with respect to the two radiation images; and apply the fat percentage Rf to the subtraction processing with respect to another image that is a radiation image in which the subject H is imaged with different timing.SELECTED DRAWING: Figure 14

Description

Technical Field

[0001] The present invention relates to a radiation image processing apparatus that provides an energy subtraction function, an operating method thereof, and a radiation image processing program.

Background Art

[0002] In the medical field, in radiation imaging for imaging a subject using radiation such as X-rays, two images with different X-ray energies are acquired from the same imaging site, and an energy subtraction process is executed to calculate the difference by utilizing the difference in X-ray energy characteristics between the bone part and the soft part, thereby generating a bone part image and a soft part image. As a result, for example, a value reflecting bone mass can be obtained from the pixel values of the created bone part image.

[0003] In bone diseases such as osteoporosis, the DXA method (Dual X-ray Absorptiometry) is known as one of the typical bone mineral quantification methods used for diagnosing bone density. The DXA method utilizes the fact that the radiation incident on and transmitted through the human body undergoes attenuation characterized by an attenuation coefficient μ (cm 2 / g) that depends on the substance (e.g., bone) constituting the human body, its density (g / cm 3 ), and thickness t (cm), and is a technique for calculating bone density from the pixel values of radiation images obtained by imaging with radiation of two types of energies. Also, various techniques for evaluating bone density using radiation images acquired by imaging a subject are known.

[0004] Specifically, in Patent Document 1, a bone part image in which the bone part of the subject is extracted is generated from a plurality of radiation images obtained by radiation with different energy distributions that have passed through the subject, the density of the region without bone in the radiation image is calculated as a correction value representing the fat content, and the pixel values of the bone part image are corrected by the correction value. Thereby, a technique for more accurately calculating the density of the bone part region and further bone density has been proposed.

Prior Art Documents

Patent Documents

[0005] [Patent Document 1] Japanese Patent Publication No. 2015-019789 [Overview of the project] [Problems that the invention aims to solve]

[0006] However, in Patent Document 1, the energy subtraction process takes longer than normal image processing, so it is necessary to wait for the energy subtraction process to finish before taking the next image. Also, since set shots are often taken in multiple directions such as frontal, lateral, and oblique views, the technician wants to change the patient's position and take images one after another without making the patient wait.

[0007] The present invention aims to provide a radiation image processing apparatus, its operating method, and a radiation image processing program that can shorten the total energy subtraction processing time and improve and stabilize the accuracy of determining the muscle-to-fat ratio in set imaging of the same person. [Means for solving the problem]

[0008] The radiation image processing apparatus of the present invention comprises a processor, which acquires at least two radiation images of the same subject using multiple different radiation energies, calculates the fat percentage of the subject's soft tissue by subtraction processing of the two radiation images, and applies the fat percentage to subtraction processing of another image, which is a radiation image of the subject taken at a different time than the radiation images.

[0009] Subtraction processing radiation It is preferable to estimate and remove scattered radiation from each pixel of the image according to the thickness distribution.

[0010] It is preferable to acquire a separate image taken from a different direction than the radiographic image, which captures the subject from a different angle.

[0011] It is preferable to use the fat percentage to correct the bone density of the subject, which is calculated from the bone region image of the radiographic image extracted by subtraction processing.

[0012] It is preferable to convert the body fat percentage to a standard soft tissue thickness and then calculate a correction factor for calculating the bone density of the subject from the standard soft tissue thickness.

[0013] It is preferable to calculate the body fat percentage from the entire image of the radiographic image.

[0014] Subtraction processing release It is preferable to extract soft tissue images from the radiation images, derive the muscle thickness and fat thickness of the subject from the soft tissue images, and calculate the fat percentage based on the fat thickness and muscle thickness.

[0015] It is preferable to derive the fat percentage from the first image taken among a series of radiographic images and apply that fat percentage to the subtraction process for the radiographic images taken thereafter.

[0016] It is preferable to apply the fat percentage to the subtraction process for the radiographic video.

[0017] It is preferable to apply a subtraction process to the radiographic images using an average fat percentage, which is the average of two or more fat percentages calculated from radiographic images of the subject taken at different times.

[0018] It is preferable to calculate the average fat percentage from the fat percentage of radiographic images taken in two different directions from among frontal, lateral, and oblique views.

[0019] It is preferable to calculate the average fat percentage from the fat percentage of radiographic images taken in three or more different directions from among a frontal view, two types of lateral views, and multiple oblique views.

[0020] It is preferable to use the muscle percentage as the proportion of soft tissue in the subject, rather than the fat percentage.

[0021] The method of operating a radiation image processing apparatus according to the present invention includes steps of: acquiring at least two radiation images each using a plurality of mutually different radiation energies for the same subject; calculating a fat percentage in the soft tissue of the subject by subtraction processing for the two radiation images; and applying the fat percentage to subtraction processing for a different image which is a radiation image taken of the subject at a timing different from that of the radiation images.

[0022] The radiation image processing program according to the present invention causes a computer to execute functions of: acquiring at least two radiation images each using a plurality of mutually different radiation energies for the same subject; calculating a fat percentage in the soft tissue of the subject by subtraction processing for the two radiation images; and applying the fat percentage to subtraction processing for a different image which is a radiation image taken of the subject at a timing different from that of the radiation images.

Advantages of the Invention

[0023] According to the present invention, in the case of shooting the same set of persons, it is possible to shorten the total time of subtraction processing, and improve and stabilize the accuracy of obtaining the ratio of muscle and fat.

Brief Description of the Drawings

[0024] [Figure 1] It is an explanatory diagram showing an outline of a radiation image processing system. [Figure 2] It is a block diagram showing functions of a radiation imaging apparatus and a radiation image processing apparatus. [Figure 3] It is an explanatory diagram of two radiation images taken with mutually different radiation energies. [Figure 4] It is a block diagram showing functions realized by an image processing unit in a radiation image processing apparatus. [Figure 5] It is an explanatory diagram of a first enhancement process for extracting a bone image and a second enhancement process for extracting a soft tissue image using the bone image. [Figure 6] This diagram illustrates the extraction of bone and soft tissue images using first and second enhancement processes, which are ES (Electron Stimulation) processing techniques. [Figure 7] These are schematic diagrams of the bony and soft tissue images of the chest region of the subject, displayed on the display 33. [Figure 8] This is an explanatory diagram showing that the contrast between bone and soft tissue in the same subject varies depending on the subject's body thickness and the tube voltage of the radiation source. [Figure 9] This is an explanatory diagram showing the relationship between the radiation attenuation coefficient and the body fat percentage of a subject when the body thickness is constant. [Figure 10] This is an explanatory diagram showing the relationship between the radiation attenuation coefficient and the body thickness of the subject when the body fat percentage is constant. [Figure 11] This is an explanatory diagram of the first lookup table, which derives a body thickness conversion coefficient that converts the body thickness of a subject to the standard soft tissue thickness based on body fat percentage. [Figure 12] This is an explanatory diagram of the second lookup table, which derives a bone density correction coefficient based on the standard soft tissue thickness with a constant fat percentage. [Figure 13] This is a block diagram showing the functions realized by the fat percentage application unit in a radiation image processing device. [Figure 14] This is an explanatory diagram showing how to apply the body fat percentage calculated by energy subtraction processing to another image of the same subject. [Figure 15] This flowchart shows the sequence of steps in the present invention. [Figure 16] This is an explanatory diagram of the function of applying body fat percentage to sequentially captured images realized in the second embodiment. [Figure 17] This is an explanatory diagram of the function realized in the third embodiment, which calculates the average body fat percentage from multiple different radiographic images of the same subject. [Modes for carrying out the invention]

[0025] [First Embodiment] An example of the basic configuration of the present invention will be described. The radiation image processing apparatus of the present invention is a personal computer or workstation or the like, on which an application program for realizing a predetermined function is installed. The computer is equipped with a processor, a CPU (Central Processing Unit), memory, and storage, and various functions are realized by the program stored in the storage. Embodiments of this disclosure will be described below with reference to the drawings.

[0026] Figure 1 is a schematic diagram of a radiation image processing system 10, which includes a radiation image acquisition device 11 (see Figure 2) and a radiation image processing device 12. The radiation image processing system 10 of this embodiment has a radiation image acquisition device 11 and a radiation image processing device 12. The radiation image acquisition device 11 and the radiation image processing device 12 are electrically connected and can transmit and receive data.

[0027] The radiographic imaging apparatus 11 comprises a radiation source 14, a radiographic panel 15, and a console 19. The radiation source 14 and the radiographic panel 15 are electrically connected to the console 19. The radiographic panel 15 has, in order from the side closest to the radiation source 14, a first radiation detector 16, a radiation energy conversion filter 17 made of a copper plate or the like, and a second radiation detector 18.

[0028] The radiographic imaging device 11 irradiates the first radiation detector 16 and the second radiation detector 18 in the radiographic imaging panel 15 with radiation Ra, such as X-rays, emitted by driving the radiation source 14 and passing through the subject H. Because the low-energy component of radiation Ra is absorbed by the radiation energy conversion filter 17, the energy of radiation Ra changes after passing through the first radiation detector 16 and before reaching the second radiation detector 18. Since the first radiation detector 16 and the second radiation detector 18 are irradiated with different energies, the radiographic imaging device 11 is an imaging device for energy subtraction using the so-called one-shot method, which acquires two radiographic images with different radiation Ra energies in a single shot. The first and second radiation detectors 16 and 18 and the radiation energy conversion filter 17 are in close contact.

[0029] The first radiation detector 16 acquires a first radiation image G1 of subject H using low-energy radiation, including so-called soft rays. The second radiation detector 18 acquires a second radiation image G2 of subject H using high-energy radiation, from which the soft rays have been removed. The first and second radiation images G1 and G2 are input to the radiation image processing device 12.

[0030] The first and second radiation detectors 16 and 18 are capable of repeatedly recording and reading out radiation images. They may be so-called direct-type radiation detectors that generate an electric charge by directly receiving radiation Ra, or they may be so-called indirect-type radiation detectors that first convert radiation Ra 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.

[0031] Figure 2 is a block diagram showing the functional configuration of the radiation image processing system 10 in this embodiment. The console 19 includes a display 20, an operation unit 21, and a communication unit 22. The radiation image processing device 12 of the present invention includes an image acquisition unit 30, an image processing unit 31, an output control unit 32, a display 33, an input receiving unit 34, and a storage memory 35.

[0032] The radiographic imaging device 11 acquires at least two radiographic images of the same subject H using energy subtraction imaging, each using two different types of radiation energy. Radiographic images with different energy distributions of radiation Ra (hereinafter referred to as energy) are distinguished as a low-energy image and a high-energy image from the two radiographic images. Since the contrast between bone and soft tissue is clear in the low-energy radiographic image, it is preferable to extract the bone image Gb, described later, from the low-energy radiographic image during energy subtraction processing. The imaging conditions can be set by input from the control unit 21 by the operator. The same subject H refers to the two images taken of the same subject H. First and second This refers to the same person and the same body part being photographed in both radiographic images G1 and G2.

[0033] Imaging using two different types of radiation energies means that the radiation Ra quality (energy) is substantially different when forming radiation images or detecting radiation Ra at least twice. The radiation Ra detected by the first radiation detector 16 and the radiation Ra detected by the second radiation detector 18 via the radiation energy conversion filter 17 will be two different types of radiation energies. As a result, the two resulting radiation images will be radiation images using different radiation energies.

[0034] The first radiation image G1 acquired from the first radiation detector 16 uses lower energy for generating the radiation image than the second radiation image G2 acquired from the second radiation detector 18. release When acquiring radiation images, two radiation images are selected from three or more distinct radiation images. The radiation image with lower energy is designated as the first radiation image G1, and the other is designated as the second radiation image G2.

[0035] As shown in Figure 3, the captured first and second radiographic images G1 and G2 can be viewed on the display 20 of the console 19. The first and second radiographic images G1 and G2 are radiographic images of the frontal view of the chest of subject H. The photographer checks the display on the display 20, and if there are no problems, operates the operation unit 21 to send the radiographic image processing device 12 from the communication unit 22. First and second The radiographic images G1 and G2 are transmitted. If retakes or multiple scans are required, the operator should input instructions from the control unit 21.

[0036] In the radiographic image processing device 12, the CPU functions as an image acquisition unit 30, an image processing unit 31, an output control unit 32, and an input receiving unit 34 by executing a radiographic image processing program. The image acquisition unit 30 acquires a first radiographic image G1 and a second radiographic image G2 from the radiographic imaging device 11. The image processing unit 31 extracts bone and soft tissue images from the first radiographic image G1 and the second radiographic image G2, derives the fat percentage, and derives bone density using the fat percentage. The output control unit 32 controls the display of each acquired image and derived data on the display 33, and transmission to the storage memory 35 or an external server.

[0037] The display 33 is a display device that displays radiation images acquired by the radiation image processing device 12. The input receiving unit 34 accepts input to the radiation image processing device 12 from input devices such as a keyboard or mouse. The connection of the display 33 and input devices is not limited to those built into or directly connected to the radiation image processing device 12, but may also be via various networks. Therefore, in the radiation image processing device 12, the display 33 and input devices may be located separately from the computer that constitutes the radiation image processing device 12. The storage memory 35 temporarily stores data used for image processing, etc., and stores acquired images and values.

[0038] The image acquisition unit 30 receives data from the radiation imaging device 11. First and second Acquire radiographic images G1 and G2. First and second The radiographic images G1 and G2 are preferably so-called original images (images that have not undergone any image processing).

[0039] The acquired first and second radiation images G1 and G2 are linked as image information to the imaging conditions, including the imaging dose (mAs value), beam quality, tube voltage, SID (Source Image receptor Distance), which is the distance between the radiation source 14 and the surfaces of the first and second radiation detectors 16 and 18, SOD (Source Object Distance), which is the distance between the radiation source 14 and the surface of the subject H, and the presence or absence of a scatter removal grid. SOD and SID are used to calculate the body thickness distribution. For SOD, it is preferable to acquire it using, for example, a TOF (Time Of Flight) camera. For SID, it is preferable to acquire it using, for example, a potentiometer, ultrasonic rangefinder, and laser rangefinder.

[0040] The body thickness distribution is preferably obtained by subtracting SOD from SID. The body thickness distribution is calculated pixel by pixel corresponding to the first and second radiation images G1 and G2. Alternatively, a method for calculating the body thickness distribution from at least one of the first and second radiation images G1 and G2, or a method for calculating the body thickness distribution from the soft tissue image of subject H described later, may be used. The body thickness distribution with the effect of scattered radiation components removed is obtained by the scattered radiation removal process described later.

[0041] Figure 4 shows the functional configuration of the image processing unit 31 of the radiation image processing device 12. The image processing unit 31 further implements the functions of a scattered radiation removal unit 40, an image extraction unit 41, a first derivation unit 42, a conversion unit 43, a second derivation unit 44, and a fat percentage application unit 45. The scattered radiation removal unit 40 removes scattered radiation from the first and second radiation images G1 and G2 based on the estimated or measured body thickness distribution. The image extraction unit 41 extracts bone and soft tissue images from the first and second radiation images G1 and G2 from which scattered radiation has been removed. The first derivation unit 42 derives the fat percentage Rf of the subject H. The conversion unit 43 converts the fat percentage Rf and body thickness T of the subject H into a standard soft tissue thickness T1. The second derivation unit 44 calculates the bone density of the subject H by performing a correction according to the fat percentage Rf. The fat percentage application unit 45 applies the derived fat percentage Rf of the subject H to another radiation image of the same subject H.

[0042] The scattered radiation removal unit 40 performs scattered radiation removal processing to remove scattered radiation components contained in the first radiation image G1 and the second radiation image G2 acquired by the image acquisition unit 30. The scattered radiation removal processing estimates and removes scattered radiation components for each pixel of the radiation image according to the body thickness, which is the estimated or measured thickness of the subject. The scattered radiation removal processing for the first radiation image G1 is described below. The scattered radiation removal processing is also performed on the second radiation image G2 in the same manner as the first radiation image G1. 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.

[0043] First, the scattered radiation removal unit 40 acquires an initial body thickness distribution Ts(x,y) as the estimated or measured body thickness. The initial body thickness distribution Ts(x,y) can be estimated from, for example, a virtual model of the subject H. The virtual model is data that virtually represents the subject H, in which the body thickness according to the initial body 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 the initial body thickness distribution Ts(x,y) is assumed to be stored in the storage memory 35 in advance, but the virtual model may be acquired from an external server. x and y are the coordinates of each pixel in the image.

[0044] When a virtual model is used, the scattered radiation removal unit 40 derives an estimated primary radiation image obtained by estimating the primary radiation image obtained by capturing the virtual model, and an estimated scattered radiation image obtained by estimating the scattered radiation image obtained by capturing the virtual model, and synthesizes the estimated image obtained by estimating the first radiation image G1.

[0045] Next, the scattered radiation removal unit 40 derives the scattered radiation component Is(x,y) and updates the body thickness distribution T(x,y) based on the difference between the estimated image and the first radiation image G1. For the detailed method, known methods may be used. The scattered radiation removal unit 40 subtracts the derived scattered radiation component (x,y) from the first radiation image G1. This removes the scattered radiation component contained in the first radiation image G1. Furthermore, the updated body thickness distribution T(x,y) obtained as a result of the removal of the scattered radiation component is obtained and can be used as the body thickness T of the subject H.

[0046] By performing the energy subtraction process described later using the first and second radiographic images G1 and G2 from which scattered radiation components have been removed, bone density can be derived with greater accuracy.

[0047] The image extraction unit 41 extracts bone and soft tissue images from the first radiation image G1 and the second radiation image G2 from which scattered radiation components have been removed. Two enhancement processes, a first enhancement process and a second enhancement process, are performed to remove the soft tissue region from the radiation image and enhance the bone region, and to obtain a bone image in which the bone region has been removed from the radiation image and the soft tissue region has been enhanced. Enhancement processing is a process that enhances structures, edges, etc., contained in the radiation image through image processing, and is a subtraction process or energy subtraction process that removes specific structures, noise, etc., from the radiation image. Alternatively, the body thickness T of the subject H may be measured from the radiation image from which scattered radiation has been removed or from the extracted soft tissue image Gs, rather than from the difference calculated in the scattered radiation removal process.

[0048] As shown in Figure 5, for example, in the image extraction unit 41, a bone image Gb is extracted from the first radiographic image G1 and the second radiographic image G2 by a first enhancement process, and a soft tissue image Gs is extracted from the first radiographic image G1 or the second radiographic image G2 using the bone image Gb by a second enhancement process. The radiographic image used as the source for extracting the soft tissue image Gs in the second enhancement process can be either the first radiographic image G1 or the second radiographic image G2.

[0049] The first enhancement process is preferably an energy subtraction process (hereinafter referred to as ES (Energy Subtraction) process). In this embodiment, ES processing is performed on two radiographic images, a first radiographic image G1 and a second radiographic image G2, using a calculation formula. In ES processing, a weighting calculation is performed in which one of the two radiographic images is weighted and subtracted from the other image. Depending on the parameters which are the weighting coefficients, signals from specific tissues such as bone or soft tissue can be reduced in the processed image after calculation processing.

[0050] As the first enhancement process, the bone image Gb is extracted by ES processing. The image generated by ES processing is called an ES image. The bone image is an ES image. In ES processing, the bone image Gb is extracted using the following equation (1) with two radiographic images, the first radiographic image G1 and the second radiographic image G2. Note that α1 is the weighting coefficient. Gb(x,y)=G1(x,y)-α1×G2(x,y) ···(1)

[0051] The second enhancement process preferably extracts the soft tissue image Gs by subtraction. To extract the soft tissue image Gs, a process is performed to remove the bone region from the first radiographic image G1 using the bone region image Gb extracted in the first enhancement process (see equation (2)), or a process is performed to remove the bone region from the second radiographic image G2 using the bone region image Gb (see equation (3)). β1 and β2 are weights. attach These are coefficients and the weights of equation (1). attach The coefficients α1 and other factors are all independent of each other. Gs(x,y)=G1(x,y)-β1×Gb(x,y) ···(2) Gs(x,y)=G2(x,y)-β2×Gb(x,y) ···(3)

[0052] Alternatively, the soft tissue image Gs may be extracted in the first weighting process, which is an ES process, and the bone tissue image Gb may be extracted using the soft tissue image Gs in the second weighting process. In that case, instead of equation (1), the soft tissue image Gs, in which the soft tissue of subject H is extracted, is generated from the first and second radiographic images G1 and G2 using equation (4) below. In the extraction of the bone tissue image Gb by the second weighting process, the relationship between the bone tissue image Gb and the soft tissue image Gs in equations (2) and (3) is reversed. In that case, the weighting coefficients will also be different values. α2 is the weighting coefficient. Gs(x,y)=G1(x,y)-α2×G2(x,y) ···(4)

[0053] As shown in Figure 6, ES processing is performed in both the first and second enhancement processes, and the first radiographic image G1 and 2nd Alternatively, a method may be used in which the bone image Gb is extracted independently using the radiographic image G2 by the first weighting process, and the soft tissue image Gs is extracted using the second weighting process. In this case, equations (1) and (4) above are used individually. In this case, the first and second weighting processes do not need to be performed in order; they can be performed simultaneously, or one weighting process can be interrupted while the other is executed.

[0054] As shown in Figure 7, it is preferable to display the bone image Gb and soft tissue image Gs extracted by the first and second enhancement processes on the display 33. The photographer (or operator) can confirm the extracted bone image Gb and soft tissue image Gs. The display may show the bone image Gb and soft tissue image Gs simultaneously when the second enhancement process is completed, or it may show the extracted images first when the first enhancement process is completed. The ES process may be repeated depending on the confirmation results. If the process is repeated, it is preferable to change the settings and methods in the extraction process.

[0055] In this embodiment, bone density in the bone region of subject H is calculated based on the bone image Gb extracted by ES processing. However, the pixel values ​​(x,y) of the bone image Gb are affected by beam hardening, and the effect of beam hardening increases as the body thickness of subject H increases. Furthermore, even with the same body thickness, the effect of beam hardening increases as the fat percentage Rf in the soft tissue region of subject H increases.

[0056] As shown in Figure 8, the higher the tube voltage, the smaller the contrast (i.e., the difference in pixel values). Figure 8 shows the relationship between the contrast between the bony and soft tissue regions of the subject H and the body thickness at three tube voltages: 80kV, 90kV, and 100kV. The higher the tube voltage (kV value) at the radiation source 14 and the higher the energy of the irradiated radiation Ra, the more the low-energy component of the radiation Ra is absorbed by the subject H during irradiation. As a result, the detected radiation Ra becomes higher energy, and the contrast between the soft tissue and bony regions of the radiation image decreases.

[0057] Furthermore, when the body thickness T of the subject H exceeds a certain value, the contrast decreases as the value of body thickness T increases. On the other hand, the contrast between the bony region and the soft tissue region becomes clearer as the fat percentage Rf increases, because the proportion of transmitted radiation energy Ra, which is relatively low energy, increases.

[0058] As shown in Figure 9, the attenuation coefficient of radiation Ra used in ES processing decreases monotonically with increasing fat percentage Rf when body thickness is constant. Also, as shown in Figure 10, when fat percentage is constant, the attenuation coefficient of radiation Ra decreases as body thickness T increases. Therefore, the attenuation coefficient of the soft tissue of subject H depends on the fat percentage Rf and body thickness T in subject H, and the effect of beam hardening of radiation Ra changes according to the fat percentage Rf and body thickness T. For this reason, using the fat percentage Rf, a correction coefficient is derived that reduces the pixel value Gb(x,y) of the extracted bone image as the fat percentage Rf increases, and bone density is corrected.

[0059] Furthermore, bone density is derived by correcting it to a value corresponding to the subject's body thickness T using a phantom. The fat percentage Rf of the material corresponding to soft tissue in the phantom is a constant value. Therefore, if only the pixel value Gb(x,y) of the bone image Gb is corrected according to the body thickness T, the bone density derived will differ from the actual bone density depending on the magnitude of the calculated original fat percentage Rf.

[0060] However, calculating bone density by correcting for both body fat percentage and body thickness increases the computational load. Therefore, bone density is derived by correcting for the effects of the body thickness T and body fat percentage Rf of subject H without directly using the body fat percentage Rf. The body fat percentage Rf and body thickness T of subject H are converted to a standard soft tissue thickness T1 with a constant body fat percentage in the conversion unit 43, and the bone density of subject H corrected using the standard soft tissue thickness T1 is derived in the second derivation unit 44.

[0061] Standard soft tissue refers to soft tissue having an attenuation coefficient and fat percentage or muscle percentage equivalent to standard human soft tissue. A phantom simulating the human body, containing a material equivalent to standard human soft tissue, can be used as the standard soft tissue. This material can be acrylic or urethane, etc. The correction coefficient for correcting the pixel value Gb(x,y) of the bone image Gb to bone density is obtained from the material equivalent to the standard soft tissue and can be derived using radiographic images of phantoms of various thicknesses. The radiation attenuation coefficient and fat percentage Rf in the phantom are predetermined values ​​corresponding to the material equivalent to the soft tissue.

[0062] The first derivation unit 42 calculates the fat percentage Rf for the entire image, that is, the average fat percentage Rf of each pixel in the soft tissue image Gs. The fat percentage Rf refers to the proportion of fat in the soft tissue of the human body. When the fat percentage of each pixel (x,y) is calculated, local variations occur due to noise, etc., so as shown in equation (5) below, the fat thickness as a percentage of the body thickness T of the subject H is calculated. R Using f(x,y), the average body fat percentage Rf is calculated for the entire image. The fat thickness tf(x,y) for each pixel is calculated from the body thickness distribution T(x,y) and body fat percentage rf(x,y) of the subject H for the corresponding pixel, and the muscle thickness tm(x,y) for each pixel is calculated from the body thickness distribution T(x,y) and muscle percentage Rm(x,y) of the subject H for the corresponding pixel. Note that body thickness T can be treated as the sum of fat thickness tf and muscle thickness tm. Rf=Σ(tf(x,y)) / Σ(tm(x,y)+tf(x,y)) ···(5)

[0063] In this embodiment, muscle tissue is approximated and derived from non-fatty tissue. The soft tissues of the human body include muscle tissue, fatty tissue, internal organs, blood, and water. Since muscle tissue, internal organs, blood, and water have similar radiation absorption characteristics, non-fatty tissue, which includes muscle tissue, internal organs, blood, and water, can be treated as muscle tissue. That is, if one of the fat percentage Rf(x, y) or muscle percentage Rm(x, y) for each pixel can be derived, the other can also be derived. Therefore, the ratio of muscle to fat in soft tissue refers to the fat percentage or the muscle percentage.

[0064] The conversion unit 43 refers to the first lookup table LUT1 and obtains a body thickness conversion coefficient K corresponding to the fat percentage Rf of the entire image derived by the first derivation unit 42. Using the obtained body thickness conversion coefficient K, the fat percentage Rf and body thickness T of the entire image of subject H are converted to a body thickness T1 of a standard soft tissue where the transmitted radiation dose Ra is equal and the fat percentage is constant. The converted body thickness T1 of the standard soft tissue is transmitted to the second derivation unit 44.

[0065] As shown in Figure 11, a body thickness conversion coefficient K is obtained using the first lookup table LUT1 to convert the body thickness T of the subject H, corresponding to the fat percentage Rf in the soft tissue image Gs, to the body thickness T1 of a standard soft tissue with a constant fat percentage. The lower the fat percentage Rf, the smaller the body thickness conversion coefficient K becomes. The first lookup table LUT1 is stored in the storage memory 35 in advance, and the conversion unit 43 refers to it in accordance with the acquisition of the fat percentage Rf from the first derivation unit 42.

[0066] The greater the thickness of the subject, the smaller the attenuation coefficient, resulting in a smaller contrast between the bony and soft tissue regions. Even if the actual bone density is the same, the greater the thickness T of the subject, the smaller the bone density value derived from the radiographic image. The second lookup table LUT2 derives the pixel value Gb(x,y) of the bony region corresponding to the body thickness T1 from radiographic images obtained by photographing phantoms of various thicknesses in advance, and derives a correction coefficient corresponding to the body thickness T1 so that the derived pixel value corresponds to the same bone density. Furthermore, the correction coefficient is derived according to various tube voltages.

[0067] The second derivation unit 44 refers to the second lookup table LUT2 and calculates a bone density correction coefficient C corresponding to the standard soft tissue thickness T1 converted by the conversion unit 43. Using the correction coefficient C, the bone density corrected for the body fat percentage Rf and body thickness T of subject H is derived from the bone image Gb. The derived bone density of subject H is saved to the storage memory 35 or displayed on the display 33 together with the bone image Gb.

[0068] As shown in Figure 12, the second lookup table LUT2 acquires a correction coefficient C for the bone image Gb, which compensates for the difference in contrast depending on the tube voltage during imaging and the decrease in contrast due to beam hardening. The second lookup table LUT2 maintains the relationship between the correction coefficient C for bone density B per pixel and the standard soft tissue thickness T1, depending on the imaging conditions, including the tube voltage setting. The larger the tube voltage and the larger the subject's body thickness, the larger the correction coefficient C becomes. The second lookup table LUT2 is stored in the storage memory 35 in advance, and the second derivation unit 44 references it when acquiring the standard soft tissue thickness T1 from the conversion unit 43.

[0069] In Figure 12, the tube voltage is 80kΩ. V This shows the relationship between body thickness and the correction coefficient when set to 90kV and 100kV. As a reference imaging condition, the correction coefficient is 1 when the tube voltage is 90kV and the body thickness is 0. Also, although the second lookup table LUT2 is shown in two dimensions, the correction coefficient C differs depending on the pixel value of the bone region. Therefore, the second lookup table LUT2 is actually a three-dimensional table with an axis representing the pixel value of the bone region. 。

[0070] The second derivation unit 44 derives the bone density B in the bone region of the subject H based on the derived correction coefficient C and the generated bone region image Gb. As shown in equation (6) below, the bone density B(x,y) for each pixel of the bone region in the bone region image Gb is derived by multiplying each pixel value Gb(x,y) by the correction coefficient C. This derives a bone density image where the pixel values ​​are bone density B(x,y). The bone density B(x,y) is obtained with the effects of beam hardening removed. done This represents the pixel value of the bone region in the radiographic image. In this embodiment, the unit of bone density is g / cm³. 2 It is assumed that this is the case. It is preferable to store the calculated bone density in conjunction with bone image Gb or radiographic image. B(x,y) = C × Gb(x,y) ... (6)

[0071] In the second derivation unit 44, a representative value of the bone density of the target area may be derived and the derived representative value of the bone density may be displayed. As the representative value, the mean, median, maximum, and minimum values ​​of the bone density can be used. Therefore, similar to the above embodiment that uses body fat percentage, a correction coefficient C corresponding to the derived body thickness T1(x,y) can be derived, and the bone density B(x,y) can be derived using the above formula (6).

[0072] Furthermore, in this embodiment, bone density is derived based on bone image Gb and fat percentage Rf, but bone density correction may be performed using muscle percentage Rm instead of fat percentage Rf. In order to approximate non-adipose tissue as muscle tissue, fat The muscle percentage Rm(x,y) can be derived from the fat percentage Rf(x,y) using the following equation (7). The muscle percentage Rm(x,y) is an example of the muscle percentage distribution of this disclosure. Rm(x,y)=1-Rf(x,y) ···(7)

[0073] Furthermore, when body thickness T is constant, the relationship between muscle mass Rm and the attenuation coefficient increases monotonically. Therefore, the relationship between muscle mass Rm and body thickness conversion coefficient K The relationship between muscle mass Rm and body thickness conversion coefficient is also monotonically increasing. K The relationship is derived and stored in memory 35, and the body thickness conversion coefficient according to the muscle ratio Rm is calculated by referring to this relationship. K By deriving this coefficient K, the thickness T of the subject H can be converted to the standard soft tissue thickness T1 using the derived thickness conversion coefficient K.

[0074] The fat percentage application unit 45 provides the fat percentage Rf of the entire image of the radiographic image, calculated by ES processing, to another radiographic image of the same subject H. Since the fat percentage of the same person is the same, the fat percentage obtained from the frontal image can be applied to another radiographic image of the same subject H. The other radiographic image may be a frontal image taken at a different time, but it may also include radiographic images of the subject H taken from a different direction than the frontal image, such as a lateral or oblique image.

[0075] As shown in Figure 13, the fat percentage application unit 45 includes the functions of a fat percentage storage unit 46, a subject determination unit 47, and a fat percentage provision unit 48. By applying the fat percentage Rf to another radiographic image, the calculation and time required for calculating the fat percentage Rf during subtraction processing can be reduced. Note that the fat percentage Rf may be calculated not only from frontal images, but also from lateral or oblique images.

[0076] The fat percentage storage unit 46 stores the fat percentage of the derived radiation image along with the information of the subject H. The fat percentage storage unit 46 may be equipped with a storage area for storage, or it may be equipped with a function to write and read data to and from the storage memory 35.

[0077] The subject determination unit 47 determines whether the subject H in the acquired radiographic image has the same body fat percentage as the saved subject H. The determination is made by comparing the patient information and shooting conditions entered during radiographic image acquisition and by calculating the correlation between the radiographic images. The orientation of the subject H during shooting is not included in the determination. If it is determined that the radiographic image was taken of the same subject H as the saved body fat percentage, a body fat percentage provision instruction is sent to the body fat percentage provision unit 48.

[0078] The fat percentage provision unit 48 applies the stored fat percentage to the acquired radiographic image in response to the received fat percentage provision instruction. The fat percentage is transmitted to the first derivation unit 42 and applied to the soft tissue image Gs before fat percentage calculation. The radiographic image to which the fat percentage has been applied is used to convert to standard soft tissue thickness and derive bone density. Note that the fat percentage provision instruction is not limited to the subject determination unit 47, but may also be given manually by the user.

[0079] As shown in Figure 14, for example, in two radiographic images in which a specific subject H is determined to be the same, the fat percentage Rf can be applied even if the source radiographic image for deriving the fat percentage is a frontal view and the destination radiographic image is a lateral view. Furthermore, the fat percentage Rf may be applied between radiographic images of the same subject H taken at different times. Even if the subject H is captured from different directions in the radiographic images, as long as the subject H is determined to be the same, the combination of imaging directions is not particularly limited.

[0080] Rather than calculating the fat percentage Rf separately for frontal and lateral radiographic images, applying the fat percentage Rf of a single radiographic image to images in different directions can improve the accuracy and stability of the fat percentage Rf. For example, if the thickness of the frontal image is smaller than that of the lateral image, the frontal image will have a relatively smaller error, thus improving the accuracy of the fat percentage Rf. Calculating the fat percentage Rf separately for the frontal and lateral images will result in different errors for each, so applying the same fat percentage Rf can stabilize the results.

[0081] The output control unit 32 displays the bone density B estimated by the second derivation unit 44 on the display 33, for example, together with the bone image Gb. Alternatively, the display 33 displays a bone density image (not shown) representing the bone density distribution in the radiographic image of the subject H, based on the derived bone density. In the bone density image, a pattern may be added to the bone region according to the magnitude of the derived bone density.

[0082] Thus, in this embodiment, the bone density in the bone region of the subject H is derived based on the bone image and fat percentage distribution of the subject H. Therefore, since the derived bone density takes into account the fat percentage Rf, the bone density can be derived with high accuracy according to this embodiment.

[0083] Furthermore, in this embodiment, as shown in Figure 1, the radiation image obtained by a simple imaging method (hereinafter referred to as simple imaging) that irradiates the subject H with radiation Ra and obtains a two-dimensional image which is a transmitted image of the subject H is used, so bone density can be easily derived. For this reason, the radiation image processing device 12 according to this embodiment can be applied to continuous use such as health checkups or monitoring the progress of treatment.

[0084] Furthermore, in this embodiment, the radiation image processing system 10 having a radiation image acquisition device 11 and a radiation image processing device 12 has been described, including the details of radiation image acquisition. However, the radiation image processing device 12 may acquire radiation images from an external storage device such as an external server, rather than from the radiation image acquisition device 11. In that case, radiation images of the same subject H that can be subjected to ES processing are acquired from the external storage device along with data such as shooting conditions, and the fat percentage is calculated. Guidance Then, image processing is performed to apply the calculated fat percentage to radiographic images of the same subject H taken from a different direction.

[0085] When acquiring radiation images from an external storage device, the image acquisition unit 30 may acquire radiation images that have undergone various processing, such as scatter correction processing or other image processing, instead of the so-called original image when acquiring two radiation images. Alternatively, the image acquisition unit 30 may be configured to ensure that at least one of the two radiation images to be acquired is a radiation image that has undergone scatter correction processing or the like.

[0086] Next, an example of the processing flow by the radiation image processing device 12 of the present invention will be explained using the flowchart shown in Figure 15. The radiation image processing device 12 acquires a first radiation image G1 and a second radiation image G2, in which the same subject H in a frontal view is captured using two different types of radiation energy (step ST110). Using the previously measured or estimated body thickness distribution T(x,y), the scattered radiation component is removed from the acquired first radiation image G1 and second radiation image G2 (step ST120). ES processing is performed on the first radiation image G1 and second radiation image G2 from which the scattered radiation component has been removed to extract a bone image Gb in which the bone portion of subject H is extracted, and a soft tissue image Gs in which the soft tissue of subject H is extracted (step ST130). From the extracted soft tissue image Gs, the fat percentage Rf of subject H in the entire image is derived (step ST140). Using the first lookup table LUT1, a body thickness conversion coefficient K corresponding to the fat percentage Rf is derived, and the body thickness T and fat percentage Rf of subject H are converted to the standard soft tissue thickness T1 using the body thickness conversion coefficient K (step ST150). Using the second lookup table LUT2, a correction coefficient C corresponding to the standard soft tissue thickness T1 is obtained, and the bone density of subject H is derived (step ST160). Images such as the bone image Gb and soft tissue image Gs, as well as calculated values ​​such as fat percentage Rf and bone density, are saved (step ST170). If there is no image processing required for the radiographic image of subject H taken from a different direction (N in step ST180), the processing of the radiographic image processing device 12 is terminated.

[0087] If there are images of the same subject H taken from different directions than the frontal view using two different types of radiation energy (Y in step ST180), the saved fat percentage Rf is applied to the ES processing of the images from the other directions, and the derivation of the fat percentage Rf is omitted (step ST190). Using the applied fat percentage Rf, body thickness conversion and bone density calculation are performed for the images from the other directions (step ST200). Images such as bone image Gb and soft tissue image Gs, as well as calculated values ​​such as fat percentage Rf and bone density, are saved (step ST210). Furthermore, if there are radiation images of the same subject H taken from different directions (Y in step ST220), the fat percentage Rf is applied when further ES processing is performed (step ST190), and bone density is calculated.

[0088] If there is no image processing required for the radiation image of subject H taken from a different direction (N in step ST220), the radiation image processing device 12 terminates processing. By applying the saved fat percentage Rf, the processing time in ES processing can be shortened, and by using the same fat percentage Rf, the calculation results of bone density and other parameters can be stabilized.

[0089] In this embodiment, the fat percentage Rf is applied to a different-angle image of the chest, using the same subject H as the subject. However, this is not limited to radiographic images of the same part of the same person. For example, radiographic images of the lumbar region capturing the pelvis or hip joint, the area around the femoral joint, or the knee may also be used.

[0090] Furthermore, in this embodiment, when performing energy subtraction processing to derive bone density, the first and second radiographic images G1 and G2 are acquired using the one-shot method, but the embodiment is not limited to this. The first and second radiographic images G1 and G2 may also be acquired using the so-called two-shot method, which involves taking two images using only one radiation detector. In the case of the two-shot method, the position of subject H included in the first radiographic image G1 and the second radiographic image G2 may shift due to the movement of subject H. For this reason, it is preferable to perform positional alignment of the subject in the first radiographic image G1 and the second radiographic image G2 before performing the processing in this embodiment.

[0091] [Second Embodiment] In the first embodiment described above, ES processing is performed on the radiographic image to obtain the fat percentage of the entire image, and the fat percentage Rf is applied to radiographic images of the same subject H taken from a different direction. In contrast, in the second embodiment, the fat percentage Rf calculated using the first radiographic image taken from a series of radiographic images is applied to the ES processing of the radiographic images taken afterward. Details common to the above embodiment will be omitted from explanation. It is preferable that the radiographic images taken consecutively are taken from the same direction.

[0092] In contrast to continuous images, which are radiographic images taken in succession, First Embodiment As shown above, the fat percentage Rf of one image is calculated, and then the same fat percentage Rf is applied to subsequent radiographic images. Applying the fat percentage Rf to consecutively captured radiographic images shortens processing time and stabilizes the results. Note that consecutive captures in consecutive images only need to be on the same day. Also, consecutive images include radiographic videos. For radiographic videos as well, the fat percentage Rf can be derived for each frame, and a single fat percentage Rf can be applied to each subsequently captured frame, similar to consecutive images.

[0093] As shown in Figure 16, when taking N consecutive radiographic images of the same subject H from the front view, the fat percentage Rf is derived from the first and second radiographic images G1 and G2 acquired in the first image and stored in the storage memory 35. The stored fat percentage Rf is then applied to the radiographic image from the second image, the third image, and so on, up to the Nth radiographic image. Since the same subject H is photographed continuously and in the same orientation, the determination of whether the subject H is the same in the subject determination unit 47 may be omitted. In that case, it is preferable to issue a fat percentage application instruction at the start of radiographic image acquisition.

[0094] [Third Embodiment] In the first and second embodiments described above, the body fat percentage Rf of the entire image, obtained from ES processing of a radiographic image taken from one direction, is applied to radiographic images of the same subject H. In contrast, in the third embodiment, the average value of the body fat percentage Rf of radiographic images of the subject H taken from multiple directions is saved and applied to radiographic images in which a specific subject H is determined to be the same. Details common to the first and second embodiments described above will be omitted from the explanation.

[0095] In the third embodiment, the body fat percentage Rf is derived from each radiation image of a specific subject H taken from multiple directions, and the average value of the derived body fat percentage Rf is calculated. The average body fat percentage Rfa is calculated from two or more body fat percentage Rfs derived from radiation images taken at different timings. It is preferable to use the body fat percentages from radiation images taken from two or more different directions for the average body fat percentage Rfa. In calculating the average body fat percentage Rfa, the body fat percentage application unit 45 implements the functions of the average body fat percentage calculation unit (not shown).

[0096] As shown in Figure 17, for example, when calculating the average fat percentage Rfa from two fat percentages Rf, The first fat percentage Rf1 is derived from a frontal radiographic image of a specific subject H and temporarily stored in the fat percentage application unit 45. Similarly, the second fat percentage Rf2 is derived from a lateral radiographic image of the same subject H and temporarily stored. The average fat percentage Rfa is calculated from the temporarily stored fat percentages Rf1 and Rf2 and stored. The average fat percentage Rfa is then applied to radiographic images of the same subject H acquired thereafter to calculate bone density. etc. To do so.

[0097] The imaging directions used to calculate the average body fat percentage (Rfa) include not only combinations of frontal and lateral views, but also combinations of any two including oblique views, and patterns using three or more types of images such as frontal, lateral, and oblique views. Furthermore, even if they are oblique views, if they are at different angles, they are treated as imaging images from different imaging directions. When n imaging images are taken with different imaging directions, the average body fat percentage calculation unit calculates the average body fat percentage (Rfa) by dividing the sum of the n body fat percentages by n. By applying the average body fat percentage (Rfa) using the body fat percentages of subjects H from multiple directions, the accuracy of the body fat percentage can be improved.

[0098] Furthermore, regarding the direction of imaging, lateral images may be distinguished as right-sided images taken from the right side and left-sided images taken from the left side, and oblique images may be captured in multiple imaging directions at any angle. For example, oblique images may be distinguished as first oblique image, second oblique image, third oblique image, etc., in order from the angle closest to the right-sided image.

[0099] In the above embodiment, in the radiation image processing device 12, picture The hardware structure of the processing unit, which executes various processes such as the image acquisition unit 30, image processing unit 31, output control unit 32, and input receiving unit 34, consists of various processors as shown below. These various processors include CPUs (Central Processing Units), which are general-purpose processors that execute software (programs) and function as various processing units; Programmable Logic Devices (PLDs), such as FPGAs (Field Programmable Gate Arrays), which are processors whose circuit configuration can be changed after manufacturing; and dedicated electrical circuits, which are processors with circuit configurations specifically designed to execute various processes.

[0100] A single processing unit may be composed of one of these various processors, or it may be composed of a combination of two or more processors of the same or different types (for example, multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, multiple processing units may be composed of a single processor. Examples of composing multiple processing units with a single processor include, firstly, a configuration where one or more CPUs and software are combined to form a single processor, and this processor functions as multiple processing units, as is typical of computers such as client and server systems. Secondly, a configuration using a processor that realizes the functions of the entire system, including multiple processing units, on a single IC (Integrated Circuit) chip, as is typical of System-on-a-Chip (SoC) systems. Thus, various processing units are configured, in terms of hardware structure, using one or more of the above-mentioned various processors.

[0101] Furthermore, the hardware structure of these various processors is, more specifically, an electrical circuit formed by combining circuit elements such as semiconductor devices. The hardware structure of the memory unit is a storage device such as an HDD (hard disk drive) or SSD (solid state drive). [Explanation of symbols]

[0102] 10. Radiation Image Processing System 11. Radiation imaging device 12. Radiation imaging processing equipment 14 Radiation source 15. Radiography Panel 16. First Radiation Detector 17. Radiation energy conversion filter 18. Second Radiation Detector 19 Console 20 displays 21 Control section 22 Communications Department 30 Image acquisition unit 31 Image Processing Unit 32 Output control unit 33 displays 34 Input Reception Section 35 Storage Memory 40 Scattered radiation removal section 41 Image Extraction Unit 42 First derivation part 43 Conversion section 44 Second derivation part 45 Fat percentage application section 46 Fat percentage storage area 47 Subject determination section 48 Fat percentage provision department C correction factor G1 First Radiation Image G2 Second Radiation Image Gb bone image Gs Soft tissue images H Subject K thickness conversion coefficient LUT1 1st Lookup Table LUT2 Second Lookup Table Ra radiation Rf fat percentage Rf1 fat percentage Rf2 fat percentage Rfa fat percentage average T Thickness T1 body thickness

Claims

1. Equipped with a processor, The aforementioned processor, For the same subject, at least two radiation images are obtained using multiple different radiation energies. Subtraction processing is performed on the two aforementioned radiation images to calculate and store the percentage of fat in the soft tissue of the subject. The thickness of the subject is obtained from another image, which is a radiographic image of the subject taken at a different timing and in a different direction than the aforementioned radiographic image. A body thickness conversion coefficient is obtained from the aforementioned stored fat percentage. The thickness of the subject is converted to the standard soft tissue thickness using the aforementioned thickness conversion coefficient. A correction coefficient for bone density corresponding to the standard soft tissue thickness is calculated. A radiographic image processing apparatus that calculates the bone density of a subject using a bone image obtained by subtraction processing on the aforementioned separate image and a bone density correction coefficient.

2. The aforementioned processor, The radiation image processing apparatus according to claim 1, wherein, for each pixel of the radiation image undergoing subtraction processing, scattered radiation is estimated and removed according to the thickness distribution.

3. The aforementioned processor, The radiation image processing apparatus according to claim 1, which calculates the fat percentage from the entire image of the radiation image.

4. The aforementioned processor, The subtraction process extracts the soft tissue image from the radiographic image, From the aforementioned soft tissue images, the muscle thickness and fat thickness of the subject are derived. The radiation image processing apparatus according to claim 3, which calculates the fat percentage based on the fat thickness and muscle thickness.

5. The radiation image processing apparatus according to claim 1, wherein the average fat percentage, which is the average of two or more fat percentages calculated from the radiation images of the subject taken at different timings, is used as the fat percentage.

6. The aforementioned processor, The radiographic image processing apparatus according to claim 5, which calculates the average fat percentage from the fat percentage of the radiographic images taken in two different directions from among a frontal image, a lateral image, and an oblique image.

7. The aforementioned processor, The radiographic image processing apparatus according to claim 5, which calculates the average fat percentage from the fat percentage of radiographic images taken in three or more different directions from among a frontal image, two types of lateral images, and multiple oblique images.

8. The aforementioned processor, A radiographic image processing apparatus according to any one of claims 1 to 7, wherein, instead of the fat percentage, the muscle percentage in the soft tissue of the subject is used to obtain the body thickness conversion coefficient, convert the body thickness of the subject to the standard soft tissue body thickness, calculate a correction coefficient for bone density corresponding to the standard soft tissue body thickness, and calculate the bone density of the subject.

9. The steps include: obtaining at least two radiation images of the same person using multiple different radiation energies; The process includes the steps of calculating and saving the percentage of fat in the soft tissue of the subject by performing subtraction processing on two of the aforementioned radiation images, The thickness of the subject is obtained from another image, which is a radiographic image of the subject taken at a different timing and in a different direction than the aforementioned radiographic image. A body thickness conversion coefficient is obtained from the aforementioned stored fat percentage. The thickness of the subject is converted to the standard soft tissue thickness using the aforementioned thickness conversion coefficient. A correction coefficient for bone density corresponding to the standard soft tissue thickness is calculated. A method for operating a radiographic image processing device that calculates the bone density of a subject using a bone image obtained by subtraction processing on the aforementioned separate image and a bone density correction coefficient.

10. A function to acquire at least two radiation images of the same person, each using multiple different radiation energies, The system includes a function to calculate and save the percentage of fat in the soft tissue of the subject by performing subtraction processing on the two aforementioned radiation images, A function to obtain the thickness of the subject from another image, which is a radiographic image of the subject taken at a different timing and in a different direction than the aforementioned radiographic image, A function to obtain a body thickness conversion coefficient from the stored fat percentage, The function of converting the thickness of the subject to the standard soft tissue thickness using the aforementioned thickness conversion coefficient, A function to calculate a bone density correction coefficient corresponding to the standard soft tissue thickness, A radiographic image processing program that causes a computer to perform a function of calculating the bone density of a subject using a bone image obtained by the subtraction process on the aforementioned separate image and a bone density correction coefficient.

Images