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

Abstract

To provide a radiation image processing device which easily and quickly generates a subtraction processing image adjusted after quantifying the image quality, an operation method of the same, and a radiation image processing program.SOLUTION: A radiation image processing device 10 generates a first emphasis image by performing first emphasis processing to two mutually different radiation images obtained by imaging a specific object Obj, generates a second emphasis image by performing second emphasis processing using at least one of the two radiation images, and generates correlation information between the first emphasis image and the second emphasis image. The device displays the first emphasis image, a level value corresponding a value of a first parameter and an image quality index in which a difference between the first emphasis image and the second emphasis image is quantified, and receives a change of the level value.SELECTED DRAWING: Figure 9

Description

Technical Field

[0005] , ,

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

Background Art

[0002] In the medical field, radiation imaging devices that image a subject using radiation such as X-rays have become widespread. For example, when the subject is a human or an animal, a radiation image taken using a radiation imaging device is used for diagnosis or treatment. In recent years, radiation images used for diagnosis and the like are not limited to so-called projection images, but also various images obtained by processing projection images are used. For example, images representing specific structures or tissues of a subject, such as a bone image showing the bone part of the subject, a soft tissue image showing the soft tissue part of the subject, and a blood vessel image showing blood vessels, are also used.

[0003] Bone images and soft tissue images can be generated by so-called subtraction processing. Subtraction processing is a process of calculating a difference by weighting at least two types of radiation images having different energies of radiation used for imaging, and is a process that utilizes the fact that the attenuation coefficient of radiation differs for each composition of the subject. Regarding subtraction processing, a method and system for energy subtraction that can be performed at low cost by using a general-purpose personal computer (PC) are known (Patent Document 1).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] When processing images generated using captured radiographic images, such as through subtraction processing, radiologists often verify the image quality of the radiographic images used for subtraction processing, and further verify and adjust the image quality of the generated processed images, in order to obtain good image quality, such as appropriately displaying the target structure or tissue. Therefore, when subtraction processing needs to be performed many times on a large number of radiographic images, such as during health checkups, it required a great deal of time and effort from radiologists.

[0006] The present invention aims to provide a radiation image processing apparatus, an operating method thereof, and a radiation image processing program that can easily and quickly generate subtraction-processed images with quantified and adjusted image quality. [Means for solving the problem]

[0007] The radiation image processing apparatus of the present invention comprises a processor. The processor acquires two radiation images taken of a specific subject using two different types of radiation energy, generates a first-enhanced image by performing a first enhancement process on the two radiation images using an arithmetic formula including a first parameter, generates a second-enhanced image by performing a second enhancement process using at least one of the two radiation images, generates correlation information showing the correlation between the first-enhanced image and the second-enhanced image, controls the display unit to display the first-enhanced image, a level value corresponding to the value of the first parameter, and an image quality index that quantifies the degree of difference between the first-enhanced image and the second-enhanced image based on the correlation information, and accepts changes to the level value.

[0008] Preferably, when the processor receives a change in the level value, it updates the value of the first parameter, generates a first enhanced image using the updated value of the first parameter, and displays the updated level value corresponding to the updated value of the first parameter on the display unit.

[0009] Preferably, the processor controls the display unit to show a user interface that accepts changes to the level value by the user.

[0010] The radiation image is preferably one in which the scattered radiation component estimated according to the thickness of the subject has been removed pixel by pixel.

[0011] The first enhancement process is preferably a subtraction process.

[0012] The subject includes both bony and soft tissue, and preferably the first weighted image is a bony image and the second weighted image is a soft tissue image.

[0013] The subject includes both bony and soft tissue, and preferably the first weighted image is a soft tissue image and the second weighted image is a bony image.

[0014] The first enhancement process preferably includes a process in which one of the two radiation images is weighted using a first parameter, and then subtracted from the other radiation image.

[0015] The second enhancement process preferably includes a process in which one of the two radiation images is weighted using the second parameter, and then subtracted from the other radiation image.

[0016] The subject includes both bone and soft tissue, the first weighted image is a bone image and the second weighted image is a soft tissue image. If, among the two radiographic images, the pixel value at coordinate (x,y) in one radiographic image is G1(x,y), the pixel value at coordinate (x,y) in the other radiographic image is G2(x,y), the pixel value at coordinate (x,y) in the bone image is Gb(x,y), and the pixel value at coordinate (x,y) in the soft tissue image is Gt(x,y), with the first parameter being α and the second parameter being β, then it is preferable that the bone image is generated by the following formula (1) and the soft tissue image is generated by the following formula (4).

[0017] Gb(x,y)=G1(x,y)-α×G2(x,y) (1) Gt(x,y) = G1(x,y) - β × G2(x,y) (4)

[0018] The second enhancement process is preferably a process of subtracting the first enhanced image from one of the two radiographic images.

[0019] The subject includes a bone part and a soft part. The first enhanced image is a bone part image, and the second enhanced image is a soft part image. Among the two radiographic images, the pixel value at the coordinates (x, y) in one radiographic image is G1(x, y), and the pixel value at the coordinates (x, y) in the other radiographic image is G2(x, y). When the pixel value at the coordinates (x, y) in the bone part image is Gb(x, y), the pixel value at the coordinates (x, y) in the soft part image is Gt(x, y), and the first parameter is α, the bone part image is preferably generated by the following formula (1), and the soft part image is preferably generated by the following formula (2) or formula (3).

[0020] Gb(x,y) = G1(x,y) - α × G2(x,y) (1) Gt(x,y) = G1(x,y) - Gb(x,y) (2) Gt(x,y) = G2(x,y) - Gb(x,y) (3)

[0021] The correlation information is preferably the correlation coefficient between the pixel values of the first enhanced image and the second enhanced image.

[0022] The image quality index is preferably the correlation coefficient.

[0023] The processor preferably sets a specific range in the image quality index as a reference range and controls to display the image quality index and the reference range on the display unit.

[0024] When the image quality index is not included in the reference range, the processor preferably performs control to notify the user.

[0025] The operation method of the radiation image processing apparatus of the present invention includes steps of: acquiring two radiation images obtained by photographing a specific subject using two different types of radiation energies respectively; generating a first enhanced image by performing a first enhancement process on the two radiation images using an arithmetic expression including a first parameter; generating a second enhanced image by performing a second enhancement process using at least one of the two radiation images; generating correlation information indicating the correlation between the first enhanced image and the second enhanced image; and performing control to display on a display unit the first enhanced image, a level value corresponding to the value of the first parameter, and an image quality index that quantifies and shows the degree of difference between the first enhanced image and the second enhanced image based on the correlation information.

[0026] The radiation image processing program of the present invention causes a computer to execute functions of: acquiring two radiation images obtained by photographing a specific subject using two different types of radiation energies respectively; generating a first enhanced image by performing a first enhancement process on the two radiation images using an arithmetic expression including a first parameter; generating a second enhanced image by performing a second enhancement process using at least one of the two radiation images; generating correlation information indicating the correlation between the first enhanced image and the second enhanced image; and performing control to display on a display unit the first enhanced image, a level value corresponding to the value of the first parameter, and an image quality index that quantifies and shows the degree of difference between the first enhanced image and the second enhanced image based on the correlation information.

Advantages of the Invention

[0027] According to the present invention, it is possible to easily and quickly generate a subtraction processed image adjusted after quantifying the image quality.

Brief Description of the Drawings

[0028] [Figure 1] It is an explanatory diagram for explaining the functions of the radiation image processing apparatus. [Figure 2] It is an explanatory diagram for explaining an example of generation of a bone part image Gb and generation of a soft part image Gt using the bone part image Gb. [Figure 3] This is an explanatory diagram illustrating an example of independently generating bone image Gb and soft tissue image Gt. [Figure 4] This is an explanatory diagram illustrating an example of generating a soft tissue image Gt and generating a bone image Gb using the soft tissue image Gt. [Figure 5] This is a lookup table showing conversion coefficients based on tube thickness and tube voltage, for tube voltages of 100kV, 90kV, or 80kV, respectively. [Figure 6] This is an explanatory diagram illustrating the display on screen 21 by the radiation image processing device. [Figure 7] This is an explanatory diagram illustrating the image quality indicators and ES level value display of ES images before the ES level value was changed. [Figure 8] This is an explanatory diagram illustrating the display of the ES image quality index and ES level value after changing the ES level value. [Figure 9] This is an explanatory diagram describing the image quality adjustment of ES images. [Figure 10] This is a flowchart explaining the processing flow of a radiation image processing device. [Figure 11] This is a schematic diagram of a radiography system. [Figure 12] This is an explanatory diagram illustrating the console's functions and connected devices. [Figure 13] This is a block diagram showing the functions of the image processing unit. [Figure 14] This is a block diagram showing the function of the body thickness distribution acquisition unit. [Figure 15] This is an explanatory diagram illustrating the display of ES image quality indicators and ES level values, showing the reference range before the ES level value was changed. [Figure 16] This is an explanatory diagram illustrating the display of ES image quality indicators and ES level values, showing the reference range after changing the ES level value. [Figure 17] This is a block diagram illustrating the functions of a radiation image processing device equipped with a notification control unit. [Figure 18] This is an explanatory diagram illustrating warning signs. [Modes for carrying out the invention]

[0029] 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 (CPU, Central Processing Unit), memory, storage, etc., and various functions are realized by the program stored in the storage.

[0030] As shown in Figure 1, the radiation image processing apparatus (hereinafter referred to as the processing apparatus) 10 of the present invention comprises an image acquisition unit 11, an image processing unit 12, a correlation information generation unit 13, a display control unit 14, a change acceptance unit 15, a display unit 16, and an operation unit 17.

[0031] The display unit 16 and the operation unit 17 may be any devices connected to the computer that constitutes the processing unit 10. The connection is not limited to a direct connection, but may also be a connection via various networks. Therefore, in the processing unit 10, the display unit 16 or the operation unit 17 may be located at a distance from the computer that constitutes the processing unit 10.

[0032] The image acquisition unit 11 acquires two radiation images of a specific subject, each taken using two different types of radiation energy. A specific subject refers to the same subject in which the captured area and direction are the same for both radiation images. Taking images using two different types of radiation energy means that when forming radiation images or detecting radiation at least twice, the radiation quality (energy distribution (hereinafter simply referred to as energy)) is substantially different for each image. This includes cases where the radiation quality of the radiation emitted by the radiation source is different for each image, as well as cases where two different types of radiation are used via a radiation energy conversion filter or the like. Therefore, the two radiation images are different.

[0033] The image acquisition unit 11 may acquire three or more radiation images when acquiring two radiation images. For example, it may acquire two different radiation images by acquiring two of three or more different radiation images taken of a specific subject using three or more different types of radiation energies.

[0034] The image acquisition unit 11 may acquire not only the so-called original image (an image without any image processing) but also radiation images that have undergone various processing, such as scattered radiation correction processing or other image processing, when acquiring two radiation images. Furthermore, the image acquisition unit 11 may be configured to ensure that at least one of the two acquired radiation images is a radiation image that has undergone scattered radiation correction processing. Scatter radiation correction processing is image processing that reduces and corrects scattered radiation from the radiation subject or other sources. Scatter radiation correction processing may also involve removing scattered radiation components estimated according to the thickness of the subject, pixel by pixel. Scatter radiation correction processing may also be performed in the image processing unit 12. Scatter radiation correction processing will be described later.

[0035] The image processing unit 12 generates a first enhanced image by performing a first enhancement process on the two radiation images acquired by the image acquisition unit 11 using a calculation formula that includes a first parameter. The image processing unit 12 also generates a second enhanced image by performing a second enhancement process using at least one of the two radiation images acquired by the image acquisition unit 11. Enhancement processing is a process that enhances structures, edges, etc., contained in the radiation image through image processing. Enhancement processing includes subtraction processing to remove specific structures, noise, etc., from the radiation image. Note that the radiation image also includes the radiation image after image processing.

[0036] In this embodiment, the first enhancement process is performed on two radiographic images, the first radiographic image G1 and the second radiographic image G2, using a calculation formula that includes a first parameter. The change reception unit 15, which will be described later, accepts changes to the level value corresponding to the value of the first parameter. When the change reception unit 15 accepts a change in the level value, the value of the first parameter is updated. The processing unit 10 repeatedly performs a series of processes each time the value of the first parameter is updated.

[0037] The first enhancement process is preferably a 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 that includes a first parameter. Specifically, the first enhanced image, which is a bone image Gb, is generated by performing ES processing on the first radiographic image G1 and the second radiographic image G2, which were acquired by the image acquisition unit 11 and which captured a specific subject as a human chest. The image generated by the ES process is called an ES image. The bone image Gb is an ES image.

[0038] In subtraction processing, a weighting operation is performed in which one of two radiographic images is weighted and subtracted from the other image. Depending on the parameter, which is the weighting coefficient, the signal from specific tissues, such as bone or soft tissue, can be reduced in the processed image. In this embodiment, subtraction processing can be performed as a process using a calculation formula that includes a first parameter α or a second parameter β, using two radiographic images, radiographic image G1 and radiographic image G2.

[0039] When performing weighted subtraction using a first parameter between corresponding pixels in two radiographic images, radiographic image G1 and radiographic image G2, for example, as shown in equation (1) below, the pixel value Gb(x,y) of a specific coordinate (x,y) in the bone image Gb is obtained by subtracting the corresponding pixel value G2(x,y) of radiographic image G2 from the corresponding pixel value G1(x,y) of radiographic image G1, weighted by the first parameter α, which is the weighting coefficient. In this case, it is preferable to set the value of the first parameter α such that the pixel value indicating soft tissue in the bone image Gb is approximately 0. Note that radiographic images G1 and G2 are images of a specific subject, and the coordinates (x,y) of the corresponding pixels in radiographic images G1 and G2 correspond to approximately the same position on the subject. Furthermore, "corresponding pixels" means pixels at positions corresponding to approximately the same position on the subject in multiple images.

[0040] Gb(x,y)=G1(x,y)-α×G2(x,y) ···(1)

[0041] When selecting which of the two radiographic images to subtract, either image can be selected. However, for the bone image Gb, it is preferable to subtract from the radiographic image with the lower radiation energy used for acquisition. Generally, the radiographic image with lower radiation energy yields an image with higher bone contrast than the radiographic image with higher radiation energy. In this embodiment, radiographic image G1 is assumed to have lower energy used for generating the radiographic image than radiographic image G2.

[0042] The first parameter α is initially set to a predetermined value. Therefore, the first parameter α is set to a predetermined value, and processing is first performed on the first radiographic image G1 and the second radiographic image G2 using a calculation formula that utilizes the predetermined value of the first parameter α. This generates the bone image Gb as the first weighted image.

[0043] The second enhancement process is a process that uses at least one of the two radiographic images. Therefore, the second enhancement process includes two types: a subtraction process using radiographic image G1 or radiographic image G2 and the first enhancement image, and a subtraction process using radiographic image G1 and radiographic image G2.

[0044] If the second enhancement process is a subtraction process, the subtraction process is the process of removing the first enhancement image from one of the two radiographic images. Removing an image means subtracting the pixel value at the corresponding pixel. The second enhancement image is generated from the pixel value resulting from the subtraction. In this embodiment, the first enhancement image is the bone image Gb. Therefore, in the second enhancement process, the bone image Gb is removed from one of the two radiographic images, radiographic image G1 and radiographic image G2, to generate the soft tissue image Gt as the second enhancement image.

[0045] Specifically, in the subtraction process, as shown in equation (2) or (3) below, the pixel value Gt(x,y) at a specific position in the soft tissue image Gt, which is the second-weighted image, is obtained by subtracting the bone image Gb(x,y), which is the first-weighted image, from the corresponding pixel value G1(x,y) of the radiographic image G1 or the corresponding pixel value G2(x,y) of the radiographic image G2.

[0046] Gt(x,y)=G1(x,y)-Gb(x,y) ···(2) Gt(x,y)=G2(x,y)-Gb(x,y) ···(3)

[0047] Therefore, when the second enhancement process is a subtraction process, the method for deriving the bone image Gb and the soft tissue image Gt is as shown in Figure 2: deriving the bone image Gb and then deriving the soft tissue image Gt by subtracting the derived bone image Gb from the original radiographic image G1 or radiographic image G2. When deriving the bone image Gb, use the following equation (1), and when deriving the soft tissue image Gt, use the following equation (2) or equation (3).

[0048] Gb(x,y)=G1(x,y)-α×G2(x,y) ···(1) Gt(x,y)=G1(x,y)-Gb(x,y) ···(2) Gt(x,y)=G2(x,y)-Gb(x,y) ···(3)

[0049] When the second enhancement process is a subtraction process, the second enhancement process is a process using an arithmetic formula that includes a second parameter applied to the two radiographic images. In this case, the second enhancement process is the same as the first enhancement process, except that the parameter is different. Therefore, it is preferable that the second enhancement process in this case is an ES process similar to the first enhancement process, and it is preferable that the process involves weighting one of the two radiographic images using the second parameter and then subtracting that weight from the other radiographic image.

[0050] For example, as shown in equation (4) below, the pixel value Gt(x,y) at a specific location in the soft tissue image Gt, which is the second-weighted image, is obtained by subtracting the corresponding pixel value G1(x,y) of the radiographic image G1, weighted by the second parameter β, from the corresponding pixel value G2(x,y) of the radiographic image G2. In this case, it is preferable to set the second parameter β such that the pixel value indicating the bone area in the soft tissue image Gt is approximately 0. Note that when the parameter α and the second parameter β are not distinguished, they are called the ES coefficient.

[0051] Gt(x,y)=G1(x,y)-β×G2(x,y) ···(4)

[0052] Therefore, when the second enhancement process is a subtraction process, the method for deriving the bone image Gb and soft tissue image Gt is as shown in Figure 3, in the first enhancement process and the second enhancement process, respectively, the bone image Gb is independently derived by the first enhancement process and the soft tissue image Gt is independently derived by the second enhancement process, using the radiographic image G1 and radiographic image G2. In the first enhancement process, when deriving the bone image Gb, the following equation (1) using the first parameter α is used, and in the second enhancement process, when deriving the soft tissue image Gt, the following equation (4) using the second parameter β is used.

[0053] Gb(x,y)=G1(x,y)-α×G2(x,y) ···(1) Gt(x,y)=G1(x,y)-β×G2(x,y) ···(4)

[0054] In the above embodiment, the bone image Gb was derived by the first enhancement process, but the soft tissue image Gt may be derived by the first enhancement process and the bone image Gt may be derived by the second enhancement process. Therefore, the soft tissue image Gt may be derived as the first enhanced image in the first enhancement process, and the bone image Gb may be derived as the second enhanced image by removing the soft tissue image Gb from one of the two radiographic images, radiographic image G1 and radiographic image G2, in the second enhancement process.

[0055] The method for deriving the soft tissue image Gt by the first enhancement process is the same as the method for deriving the bone tissue image Gb by the first enhancement process. That is, weighted subtraction using the second parameter is performed between the corresponding pixels of the two radiographic images, radiographic image G1 and radiographic image G2. This is the same as when the soft tissue image Gt was derived using equation (4) below as described above.

[0056] Gt(x,y)=G1(x,y)-β×G2(x,y) ···(4)

[0057] Then, in the subtraction process, which is the second weighting process, the pixel value Gb(x,y) at a specific position in the bone image Gb, which is the second weighting image, is obtained by subtracting the soft tissue image Gt(x,y), which is the first weighting image, from the corresponding pixel value G1(x,y) of the radiographic image G1 or the corresponding pixel value G2(x,y) of the radiographic image G2, as shown in equation (5) or (6) below.

[0058] Gb(x,y)=G1(x,y)-Gt(x,y) ···(5) Gb(x,y)=G2(x,y)-Gt(x,y) ···(6)

[0059] Therefore, when the second enhancement process is a subtraction process, the method for deriving the bone image Gb and soft tissue image Gt may be as shown in Figure 4, by deriving the soft tissue image Gt and then subtracting the derived soft tissue image Gt from the original radiographic image G1 or radiographic image G2 to derive the bone image Gb. When deriving the soft tissue image Gt, use the following equation (4), and when deriving the soft tissue image Gt, use the following equation (5) or equation (6).

[0060] Gt(x,y)=G1(x,y)-β×G2(x,y) ···(4) Gb(x,y)=G1(x,y)-Gt(x,y) ···(5) Gb(x,y)=G2(x,y)-Gt(x,y) ···(6)

[0061] In the subtraction process, when selecting which of the two radiographic images to subtract from, either image can be selected. However, it is preferable to select the image with the higher radiation energy used for acquisition, as this will result in an image with better bone removal. Generally, the image with higher radiation energy will have lower bone contrast than the image with lower radiation energy.

[0062] Regarding the parameters, predetermined values ​​are set in advance. Therefore, parameter β is set to a predetermined value in advance, and first, processing is performed on the first radiation image G1 and the second radiation image G2 using a calculation formula that utilizes parameter β with the predetermined value. This generates a soft tissue image Gt as the second weighted image. The soft tissue image Gt is an ES image.

[0063] Furthermore, if the acquired radiation image has not undergone contrast correction processing, it is preferable for the image processing unit 12 to perform contrast correction processing on the acquired radiation image to correct for the difference in contrast between the two radiation images according to the tube voltage at the time of acquisition, and for the decrease in contrast due to the effect of beam hardening. A known method can be used for the contrast correction processing.

[0064] Figure 5 shows a lookup table for obtaining correction coefficients to compensate for the contrast differences in the bone image Gb depending on the tube voltage during acquisition, and for the decrease in contrast due to beam hardening. The correction coefficient is a coefficient for correcting each pixel value Gb(x,y) in the bone image Gb. Lookup table LUT20, with the reference acquisition conditions set to a tube voltage of 90kV, is shown as an example.

[0065] As shown in Figure 5, in the lookup table LUT20, a larger correction coefficient is set for larger tube voltages and larger subject thicknesses. Note that although the lookup table LUT20 is shown in two dimensions in Figure 5, the correction coefficient differs depending on the pixel values ​​of the bone region. Therefore, the lookup table LUT20 is actually a three-dimensional table with an axis representing the pixel values ​​of the bone region.

[0066] Similarly to the bone image Gb, the soft tissue image Gt can also be corrected for fluctuations in the pixel values ​​Gt(x,y) of the soft tissue image Gt by using a lookup table (not shown) to obtain a correction coefficient corresponding to the soft tissue image Gt as a contrast correction process.

[0067] The correlation information generation unit 13 generates correlation information showing the correlation between the first-weighted image and the second-weighted image. When the correlation is low, it can be said that the first-weighted image and the second-weighted image are images that are highly different from each other, and when the correlation is high, it can be said that the first-weighted image and the second-weighted image are images that are highly similar. Therefore, the degree of difference between the first-weighted image and the second-weighted image can be quantified using the image quality index based on the correlation information. Accordingly, it is preferable to convert the correlation information into an image quality index such that the lower the correlation between the first-weighted image and the second-weighted image, the higher the image quality as an ES image, and the higher the correlation between the first-weighted image and the second-weighted image, the lower the image quality as an ES image.

[0068] Such correlation information includes correlation values. Specifically, any correlation value that indicates the degree of correlation between images can be used, whether it be a statistically calculated correlation coefficient, or the correlation values ​​used in template matching techniques such as SAD (Sum of Absolute Difference), SSD (Sum of Squared Difference), NCC (Normalized Cross-Correlation), or ZNCC (Zero-mean Normalized Cross-Correlation).

[0069] By generating correlation information between the first-weighted image and the second-weighted image as a correlation value, the degree of difference between the first-weighted image and the second-weighted image is quantified. Therefore, if it is preferable that the first-weighted image and the second-weighted image are different from each other, it is most preferable that the correlation value is a correlation value that indicates the degree of the difference to the greatest extent possible in terms of the quantified degree of difference.

[0070] In this embodiment, the correlation coefficient used in statistics is used as correlation information, which is a correlation value indicating the degree of correlation, and is calculated using the pixel values ​​of the first-weighted image and the pixel values ​​of the second-weighted image. The correlation coefficient may be calculated for the entire first-weighted image and the second-weighted image, or it may be calculated for the region of interest in the first-weighted image and the corresponding region of interest in the second-weighted image.

[0071] The correlation coefficient r between the pixel value of the region of interest in the first-weighted image and the pixel value of the corresponding region of interest in the second-weighted image can be calculated using the following example. The correlation coefficient r between the first pixel value in the region of interest in the bone image Gb (first-weighted image) and the second pixel value in the corresponding region of interest in the soft tissue image Gt (second-weighted image) is calculated using the following equation (7) from the standard deviation sa of the first pixel value, the standard deviation sb of the second pixel value, and the covariance sab between the first and second pixel values.

[0072]

number

[0073] The standard deviation sa of the first pixel value, the standard deviation sb of the second pixel value, and the covariance sab between the first and second pixel values ​​are calculated using the following equations (8), (9), or (10), respectively. The pixel value at each coordinate in the region of interest of the bone image Gb is defined as the first pixel value ai (where i is an integer), and the numerical value of the pixel at each coordinate in the region of interest of the soft tissue image Gt, which is the second-weighted image, is defined as the second pixel value bi (where i is an integer). The mean value of the first pixel value ai is defined as mean A, the mean value of the second pixel value bi is defined as mean B, and the total number of first pixel values ​​ai and second pixel values ​​bi is defined as total n.

[0074]

number

[0075]

number

[0076]

number

[0077] Therefore, the correlation coefficient r between the first pixel value ai and the second pixel value bi can be calculated using the following equation (11).

[0078]

number

[0079] The correlation coefficient r takes a value within the range of 1 to -1. The closer it is to 1, the stronger the positive correlation between the first pixel value ai and the second pixel value bi. The closer it is to -1, the stronger the negative correlation. The closer it is to 0, the less correlation there is. Since the bone image Gb and the soft tissue image Gt are generated by subtracting the image signals of either the soft tissue or bone from the two radiographic images, the closer the correlation coefficient r is to 0, the less likely it is that the first pixel value ai of the bone image Gb and the second pixel value bi of the soft tissue image Gt share similar pixel values ​​at a specific location, and the higher the degree to which they are distinct images. Therefore, if the bone image Gb and the soft tissue image Gt are distinct images to a high degree, the image quality of the bone image Gb and the soft tissue image Gt can be considered to be high. In this specification, "image quality" for ES images refers to the presence or absence of images or pixel values ​​based on structures or tissues other than the target structure in the ES image. As described above, the correlation coefficient r between the first pixel value ai of the bone image Gb and the second pixel value bi of the soft tissue image Gt is suitable as an indicator of image quality for both the bone image Gb and the soft tissue image Gt. The closer the correlation coefficient r is to 0, that is, the closer the image quality indicator is to 0, the higher the image quality of the bone image Gb or the soft tissue image Gt.

[0080] The correlation coefficient r between the first pixel value ai of the bone image Gb and the second pixel value bi of the soft tissue image Gt changes when the bone image Gb and / or soft tissue image Gt are modified. The bone image Gb is modified when the first parameter α is updated from its previous value. When the soft tissue image Gt is modified, there are two cases depending on the content of the second enhancement processing when the soft tissue image Gt was generated: the bone image Gb is modified, or the second parameter β is updated from its previous value. It is preferable for the correlation information generation unit 13 to generate correlation information each time an ES image is generated. Therefore, in this embodiment, it is preferable for the correlation information generation unit 13 to calculate the correlation coefficient r each time the bone image Gb and / or soft tissue image Gt are modified.

[0081] The display control unit 14 controls the display unit 16 to display the first-enhanced image, the level value corresponding to the value of the first parameter used when generating the first-enhanced image, and the image quality index which quantifies the degree of difference between the first-enhanced image and the second-enhanced image based on correlation information. The user of the processing unit 10 can confirm the first-enhanced image, the level value, and the image quality index displayed on the display unit 16 through the screen of the display unit 16.

[0082] The display unit 16 is, for example, a display such as a liquid crystal, and displays the captured radiation image and the processed image generated by the radiation image processing device 10. In this embodiment, the display unit 16 is a display. The operation unit 17 is, for example, a keyboard and / or a pointing device for operating the processing device 10. The display unit 16 and the operation unit 17 can be configured as touch panels. In this embodiment, the operation unit 17 is used to give various instructions such as input, selection, or change via a GUI (Graphical User Interface) displayed on the display.

[0083] The ES image quality index, which quantifies the degree of difference between the first-weighted image and the second-weighted image based on correlation information, is based on correlation information and uses a quantifiable representation of the image quality of the first-weighted image and the second-weighted image. It is preferable that the ES image quality index is such that the superiority or inferiority of the image quality of the first-weighted image or the second-weighted image can be easily understood by the value of the image quality index. Depending on the type of correlation information, the ES image quality index may be a value obtained by transforming the correlation value of the correlation information, or it may be the correlation value of the correlation information itself.

[0084] In this embodiment, the ES image quality index is the correlation coefficient r itself. By using the correlation coefficient r itself as the ES image quality index, the correlation value takes a value within the range of -1 to +1, making it easy to understand the quality of the ES images. According to such an ES image quality index, the smaller the absolute value of the ES image quality index and the closer it is to 0, the better the image quality based on the correlation information of the ES image (hereinafter referred to as ES image quality). Conversely, the higher the absolute value of the ES image quality index and the closer it is to -1 or +1, the less good the ES image quality of the ES image is, suggesting that there may be room for image quality adjustment.

[0085] As shown in Figure 6, in this embodiment, the display screen 21 shows the bone image Gb, which is the first weighted image. In addition to the bone image Gb, the screen 21 includes an image adjustment unit 22 that performs various general adjustments when generating the bone image Gb, an ES image switching unit 23 that gives instructions to switch the type of ES image or radiographic image displayed on the screen 21, an ES image quality index display unit 24 that displays the ES image quality index of the displayed bone image Gb, and an ES level display unit 25 that displays a level value corresponding to the value of parameter α used when generating the displayed bone image Gb.

[0086] The ES image switching unit 23 issues instructions to switch the type of ES image displayed on screen 21. In addition to switching ES images, the ES image switching unit 23 may also be configured to switch and display radiographic images. As shown in Figure 7, when the type of ES image includes bone image Gb and soft tissue image Gt, the type of ES image to be displayed on screen 21 is selected by selecting one of the "bone" or "soft tissue" options displayed on screen 21 using a GUI or the like. In the ES image switching unit 23, it is preferable to display which of the "bone" or "soft tissue" options has been selected according to the selection of the type of ES image, for example, by displaying the unselected option in a darker color and the selected option in a brighter color, so that it can be understood at a glance. The selected type of ES image is displayed on the left side of screen 21. In this embodiment, bone image Gb has been selected by the ES image switching unit 23, and bone image Gb is displayed on screen 21.

[0087] The ES image quality indicator display unit 24 preferably displays the ES image quality indicator in a way that allows for easy understanding. The display format of the ES image quality indicator is set according to the type of ES image indicator. The display format of the ES image quality indicator is sufficient as long as the user can understand the ES image quality indicator.

[0088] In this embodiment, the ES image quality index is the correlation coefficient r itself, and is therefore displayed in the form of a number line 31 that shows a range from -1 to 1 with a scale of 0 at the origin. The ES image quality index of the bone image Gb displayed on screen 21 is indicated by a point 32 on the number line 31. This makes it possible to grasp at a glance whether the ES image quality index of the bone image Gb displayed on screen 21 is relatively close to or far from the scale of 0, which is the most desirable for ES image quality.

[0089] The ES image quality is adjusted by regenerating the ES image. As described above, the ES image is generated by the image processing unit 12 performing processing on the two acquired radiation images using a calculation formula that includes ES coefficients. Therefore, by changing the ES coefficients used when generating the ES image and regenerating the ES image using the changed ES coefficients, an ES image with adjusted image quality can be generated.

[0090] Regarding changes to the ES coefficient, the value of the ES coefficient used to generate the ES image may be used as is, but it is preferable to convert it to an ES level value, which is a level value corresponding to the ES coefficient. The ES level value is, for example, a predetermined numerical range corresponding to the ES coefficient. In this case, the numerical range can be, for example, within the range of 0 to 10, or within the range of 0 to 100. This allows the user to easily understand the relationship between changing the value of the ES coefficient and the degree of adjustment of the ES image quality.

[0091] The ES level display unit 25 includes an ES level value display unit 34 and an ES level value changing unit 35. The ES level value display unit 34 displays the level value, which is the ES level value of the bone image Gb displayed on the screen 21, as a number. For example, the level value when the bone image Gb displayed on the screen 21 was generated is displayed as "2.1". In this embodiment, the level value is a numerical value within the range of 1 to 10, corresponding to the value of parameter α, and shown with one decimal place. The ES level value changing unit 35 is displayed by a GUI and has buttons for changing the ES level value. The buttons consist of, for example, a plus button labeled "+" and a minus button labeled "-".

[0092] The change reception unit 15 accepts changes to the ES level value. When the change reception unit 15 accepts a change to the ES level value, it updates the value of the first parameter corresponding to the ES level value. When the value of the first parameter is updated, it performs a series of processes again using the updated value of the first parameter, and controls the display unit 16 to display the new first enhanced image, level value, and image quality index, respectively. Therefore, the above series of processes is performed again each time a change to the level value is accepted.

[0093] Specifically, when the change reception unit 15 receives a change in the ES level value, the image processing unit 12 generates and updates a first enhanced image by performing a first enhancement process using the updated first parameter value, and, if necessary, performs a second enhancement process using the updated first enhanced image. The correlation information generation unit 13 and the display control unit 14 also perform processing again using the updated first enhanced image, and the display unit 16 displays the updated first enhanced image, the level value corresponding to the updated first parameter, and the updated image quality index.

[0094] In this way, each time the change reception unit 15 receives a change in the ES level value, the process from generating the processed image to displaying the image quality index, as described above, is repeated. The user can look at the image quality index displayed on the display and the newly generated first enhanced image, and may change the ES level value again, or may terminate the process if they think the ES image quality has improved.

[0095] Furthermore, in order for the change acceptance unit 15 to accept changes to the ES level value, it is preferable that the display control unit 14 controls the display of the display unit 16 to show a user interface that accepts changes to the ES level value by the user. The user interface, etc., can be any interface that can accept changes to the ES level value by the user, and its form and other aspects are not limited.

[0096] As shown in Figure 8, in this embodiment, the user changes the ES level value by operating the ES level value change unit 35 with an operation unit 17 such as a mouse or keyboard. For example, if the operation unit 17 is a mouse, an arrow-shaped cursor 36 is displayed on the screen 21, and by clicking the button selected with the cursor 36, the user can increase or decrease the changed ES level value displayed on the ES level display unit 25. In Figure 8, the value, which was previously "2.1", has been changed to "2.8". When the changed ES level value is displayed on the ES level value change unit 35, the change reception unit 15 accepts the change in the ES level value, and the subsequent processing proceeds automatically as described above. The bone image Gb that has been regenerated in accordance with the changed ES level value is displayed, and the image quality index of the regenerated bone image Gb is displayed as a point 32 on the ES image quality index display unit 24.

[0097] As shown in Figure 9, if the image quality index of the newly generated first-weighted image is far from the reference value of 0, the user can generate a first-weighted image with improved ES quality by repeatedly changing the ES level value. Each time the change reception unit 15 receives a change in the ES level value, the process from generating the processed image to displaying the image quality index as described above is repeated. Finally, when the point 32 indicating the ES image quality index of the modified bone image Gb becomes closer to the reference value of 0 for the ES image quality index, the user can see at a glance on the screen 21 that the newly generated bone image Gb has good ES quality, does not require further adjustment, and that the ES quality adjustment is complete.

[0098] Thus, even when the ES level value is changed multiple times, the changed ES image quality index is displayed each time, allowing the user to understand the trend of how much the ES image quality index improves or deteriorates depending on how the ES level value is converted. In this way, the user can adjust the ES image by performing operations in one place while viewing the ES image on a single screen 21, making it easy and quick to adjust the ES image quality to be as close as possible to the baseline of 0.

[0099] In this embodiment, the first weighted image was described as a bone image Gb, but the first weighted image may be a soft tissue image Gt and the second weighted image may be a bone image Gb.

[0100] Furthermore, in this embodiment, the radiographic image is an X-ray image, and the first-weighted image and second-weighted image are described as a bone image Gb and a soft tissue image Gt. However, any radiographic image can be used as long as it is possible to generate an image in which a specific structure or tissue is extracted by performing subtraction or other processing on the two radiographic images. For example, the radiographic image may be a CT (Computed Tomography) image, and the first-weighted image and second-weighted image may be a vascular image and an image excluding blood vessels, etc.

[0101] Next, an example of the processing flow by the processing device 10 of the present invention will be explained using the flowchart shown in Figure 10. The processing device 10 uses two different types of radiation energy to acquire a first radiation image G1 and a second radiation image G2, which are taken from the same area of ​​the chest of the same subject as a specific subject, in the same orientation (step ST110). A bone image Gb is generated by performing ES processing on the acquired first radiation image G1 and second radiation image G2 using a calculation formula that includes a first parameter. A soft tissue image Gt is generated by subtracting the bone image Gb from the second radiation image G2, and two types of ES images, the bone image Gb and the soft tissue image Gt, are generated (step ST120). Next, correlation information between the bone image Gb and the soft tissue image Gt is calculated (step ST130). The correlation information is the correlation coefficient r. Then, the value of the correlation coefficient r is used as the ES image quality index, and an image quality index is generated (step ST140).

[0102] Here, among the bone image Gb and soft tissue image Gt, the bone image Gb will be adjusted for image quality. Therefore, the display shows the bone image Gb, the ES image quality index, and the level value, which is the ES level value based on the first parameter α used in the calculation formula when generating the bone image Gb (step ST150). The radiologic technologist, as the user, decides whether to change or not change the ES level value based on the display. If the ES level value is changed (Y in step ST160), the bone image Gb is generated again by performing ES processing using a calculation formula that includes the first parameter based on the changed ES level value (step ST180). If the ES level value is not changed (N in step ST160), the image quality adjustment of the bone image Gb is considered complete (step ST170).

[0103] ES images sometimes suffered from errors due to fluctuations in radiation output values ​​or variations in the subject, resulting in degraded image quality. For example, in the generated bone image Gb, bone areas were sometimes completely obscured or partially obscured. Two possible causes for this are: firstly, changes over time in the radiation source and / or radiation imaging panel of the radiation imaging device (see Figure 11). Changes over time due to deterioration of the radiation tube in the radiation source alter the quality and / or dose of the radiation. Secondly, changes over time due to deterioration of the radiation imaging panel alter the output pixel values.

[0104] Furthermore, the correction coefficient used in LUT20 (see Figure 5) to compensate for contrast reduction relies on the set tube voltage information. Therefore, if the actual radiation quality changes, errors will occur in the correction coefficient. Also, body thickness is determined from the difference between the pixel value I0 when there is no subject and the pixel value I when there is a subject during imaging. However, the pixel value I0 is calculated based on calibration data acquired in advance using set imaging condition information such as tube voltage (kV value), dose (mAs value), and SID (Source Image receptor Distance). Therefore, if the actual radiation quality, dose, or output value of the radiography panel changes compared to when the calibration data was acquired, errors will occur in the calculated pixel value I0, causing the body thickness or correction coefficient in LUT20 to shift. As a result, errors may occur in the generated ES images such as bone images Gb.

[0105] Furthermore, when estimating primary or scattered radiation, the body thickness is repeatedly determined to match the pixel value I. Body thickness is determined from the difference between the pixel value I0 when there is no subject and the pixel value I when there is a subject. Therefore, if the radiation quality, dose, or output value of the radiography panel changes, an error will occur in the calculated pixel value I0, causing the body thickness to shift and resulting in an error in the primary or scattered radiation being determined. Consequently, since the bone image Gb is calculated using the pixel values ​​of the radiographic image, which are primary radiation data containing errors, errors may occur in the bone image Gb, which is an ES image.

[0106] Secondly, there are cases where the thickness of the subject varies. For example, in LUT20 (see Figure 5), the pixel value Gb(x,y) of the bone image Gb is corrected depending on the thickness of the subject. However, the body thickness in this relation is the body thickness at a standard fat percentage, and Gb(x,y) is corrected depending on the thickness of the subject. However, the fat percentage may vary from the standard value depending on the subject and body part. As a result, a deviation occurs in the correction coefficient of Gb(x,y), and in the ES image, bone may be over-exposed or under-exposed. Therefore, for example, a deviation occurs in the correction coefficient of the pixel value Gb(x,y) of the bone image Gb, and in the ES image of the bone image Gb, bone may be over-exposed or under-exposed.

[0107] As described above, when ES images are thought to have errors, such as excessive or incomplete disappearance of the target structure, image adjustment by a radiologic technologist was necessary. Radiologic technologists visually check the ES images and adjust the image quality. In cases requiring a large number of radiographs, such as for health checkups, radiologic technologists must complete the shooting one after another in a short time. Therefore, there may be insufficient time to check the image quality of the ES images created based on the regular radiographs, and the image quality of the ES images at the time of shooting may not be guaranteed.

[0108] To address these problems, solutions such as improving the processing procedures or processing speed of the image quality adjustment system or equipment can be considered. However, even after solving these problems, if the system relies on radiologic technologists visually evaluating radiographic or ES images to adjust the image quality, it is possible that the burden on radiologic technologists will still remain. Furthermore, in this case, there is a possibility of variability in image quality adjustment due to individual differences among radiologic technologists.

[0109] According to the processing device 10, etc., the ES image quality index, which represents the ES image quality, is quantified and displayed on the screen. Therefore, radiographers only need to check the image quality index, reducing the burden of image verification or image quality adjustment of ES images. In addition, since the quantified image quality and, in some cases, its standard are displayed, variations in visual evaluation due to individual differences among radiographers can be reduced.

[0110] Furthermore, when adjusting image quality, the ES image quality can be quickly adjusted simply by changing the ES level value with a click or similar operation, thus reducing the burden of image quality adjustment. This allows for rapid adjustment of ES image quality even during imaging. Therefore, radiologic technologists can handle the processing of a large volume of radiographic images and reduce the frequency of re-imaging due to image defects, etc.

[0111] As described above, the processing device 10 reduces the burden on radiologists to verify image quality by quantifying and displaying ES image quality. It also reduces variability in visual evaluation due to individual differences among radiologists. Furthermore, since the ES image quality index is related to the generation of ES images, adjusting ES image quality can be done simply by changing the ES level value, making image quality adjustment easy and reliable. Therefore, the processing device 10 makes it possible to easily and quickly generate ES images with better image quality. Consequently, the processing device 10 and the like significantly reduce the burden on radiologists to verify the image quality of ES images during imaging.

[0112] Next, an example of an embodiment of a radiography system including a processing unit 10 will be described. As shown in Figure 11, the radiography system 40 of this embodiment comprises a radiation source 41, a radiography panel 42, a console 43, and a processing unit 10. The radiation source 41, the radiography panel 42, and the console 43 constitute a radiography apparatus.

[0113] A radiography apparatus can capture an entire image of a specific subject. Therefore, a radiography apparatus consists of a radiation source 41 and a radiography panel 42, etc., which can capture an image of the entire chest of a specific subject, a person.

[0114] The radiation source 41 is a device that generates radiation Ra necessary for imaging, and consists of a radiation tube that generates radiation Ra and a high-voltage generation circuit that generates the high voltage necessary for the radiation tube to generate radiation Ra. By adjusting the tube voltage and tube current of the radiation tube, the radiation source 41 can generate multiple types of radiation with different beam qualities (energy distribution (hereinafter simply referred to as energy)). In this embodiment, the radiation source 41 is an X-ray source that generates X-rays. Therefore, the radiography system 40 is an X-ray imaging system that acquires an X-ray image of a subject Obj by imaging the subject Obj using X-rays. The subject Obj is, for example, a person, and in this embodiment, the chest of a person is imaged as a specific subject Obj.

[0115] The radiography panel 42 uses radiation Ra generated by the radiation source 41 to image the subject Obj. In other words, the radiography panel 42 is a so-called FPD (Flat Panel Detector) that detects radiation Ra transmitted through the subject Obj and converts it into an electrical signal to output a radiographic image of the subject Obj. When imaging using the radiography panel 42, a grid (not shown) can be used in combination as needed. A grid is a device that removes the scattered radiation component, such as a stationary Lissholm grid or a mobile Bucky grid.

[0116] In this embodiment, the radiography panel 42 includes two detectors: a first radiation detector 44 and a second radiation detector 45. Of the first and second radiation detectors 44 and 45, the first radiation detector 44 is positioned relatively close to the subject Obj and the radiation source 41, while the second radiation detector 45 is positioned relatively far from the subject Obj and the radiation source 41. The first and second radiation detectors 44 and 45 detect radiation Ra transmitted through the subject Obj pixel by pixel. The first and second radiation detectors 44 and 45 also output radiation images of the subject Obj, respectively.

[0117] However, the radiography panel 42 includes a radiation energy conversion filter 46 between the first radiation detector 44 and the second radiation detector 45. The radiation energy conversion filter 46 is, for example, a copper plate and absorbs the low-energy component of radiation Ra. Therefore, the energy of radiation Ra changes after passing through the first radiation detector 44 and before reaching the second radiation detector 45. Thus, although the radiography panel 42 simultaneously images a specific subject Obj under the same imaging conditions (same radiation Ra), the first radiation image G1 (see Figure 1) output by the first radiation detector 44 and the second radiation image G2 (see Figure 1) output by the second radiation detector 45 are essentially two different types of radiation Ra energies used to image the specific subject. This method is sometimes called one-shot energy subtraction (one-shot ES).

[0118] The first radiation detector 44 and the second radiation detector 45 may be either indirect conversion type detectors or direct conversion type detectors, and different types of detectors can be used for the first radiation detector 44 and the second radiation detector 45. An indirect conversion type detector is a detector that indirectly obtains an electrical signal by converting radiation Ra into visible light using a scintillator made of CsI (cesium iodide), and then photoelectrically converting that visible light. A direct conversion type detector is a detector that directly converts radiation Ra into an electrical signal using a scintillator made of amorphous selenium, etc. In addition, the first radiation detector 44 and the second radiation detector 45 may be either PSS (Penetration Side Sampling) type detectors or ISS (Irradiation Side Sampling) type detectors, respectively. The PSS method is a method in which the scintillator is placed on the subject Obj side relative to the TFT (Thin Film Transistor) that reads out the electrical signal. The ISS method, unlike the PSS method, is a method in which the scintillator and TFT are arranged in the order of TFT followed by scintillator from the subject (Obj) side.

[0119] Furthermore, if the radiation source 41 generates two or more types of radiation with different radiation qualities by adjusting the tube voltage and tube current of the radiation tube, the radiography panel 42 is equipped with at least one radiation detector. In this case, the radiation detector can be the same as the first radiation detector 44 or the second radiation detector 45, etc.

[0120] In this case, the radiography panel 42 takes at least two images of a specific subject Obj using different radiation qualities, under the same imaging conditions except for the difference in radiation quality. The first radiographic image G1 (see Figure 1) output by the first radiation detector and the second radiographic image G2 (see Figure 1) output by the second radiation detector are radiographic images of the specific subject taken using two different types of radiation Ra energies, respectively. This method is sometimes called two-shot energy subtraction (two-shot ES).

[0121] The console 43 is a control device (computer) that controls the operation of the radiation source 41 and the radiography panel 42, etc., and is, for example, a personal computer or workstation with an application program installed to realize a predetermined function. As shown in Figure 12, the console 43 may be connected to a RIS 47 (Radiology Information Systems), HIS 48 (Hospital Information Systems), PACS 49 (Picture Archiving and Communication Systems), or a DICOM (Digital Imaging and Communications in Medicine) server (not shown) included in PACS 49, and acquires imaging information such as imaging orders related to the acquisition of radiographic images from these, and also transmits the acquired images.

[0122] The console 43 includes a display unit 47 and an operation unit 48. The display unit 47 is, for example, a liquid crystal display, and displays the captured radiation image and other necessary information related to operation or settings. The operation unit 48 is, for example, a keyboard and / or a pointing device, used for inputting settings such as shooting conditions and operating the radiation source 41 and the radiation imaging panel 42. The display unit 47 and the operation unit 48 can be configured as touch panels.

[0123] In this embodiment, the processing unit 10 is a separate device from the console 43, but part or all of the processing unit 10 can be installed in the console 43. In this case, the display unit 16 and / or operation unit 17 of the processing unit 10 can use the display unit 47 and / or operation unit 48 of the console 43. Also, if the entire processing unit 10 is installed in the console 43, the console 43 constitutes the processing unit 10.

[0124] As described above, the processing unit 10 comprises an image acquisition unit 11, an image processing unit 12, a correlation information generation unit 13, a display control unit 14, a change acceptance unit 15, a display unit 16, and an operation unit 17 (see Figure 1). The display unit 16 may be shared with the display unit 47 of the console 43. Programs related to various processes are stored in the processing unit 10 in a program memory (not shown). In the processing unit 10, the programs in the program memory are operated by a central control unit (not shown) composed of a processor or the like, thereby realizing the functions of the image acquisition unit 11, image processing unit 12, correlation information generation unit 13, display control unit 14, change acceptance unit 15, display unit 16, operation unit 17, and the central control unit.

[0125] The processing unit 10 is directly connected to the console 43, and the image acquisition unit 11 can acquire radiographic images of the subject Obj in real time and use them for image processing. In addition to directly connecting to the console 43, the processing unit 10 may also acquire radiographic images indirectly via RIS 47, HIS 48, PACS 49, or a DICOM server (not shown) included in PACS 49 and use them for image processing. Furthermore, the processing unit 10 may transmit and store processed images such as bone images Gb or soft tissue images Gt, which have been adjusted by the processing unit 10, to an external device such as a DICOM server.

[0126] As mentioned above, when the image processing unit 12 performs scattered radiation correction processing, it may also include a body thickness distribution acquisition unit 61, as shown in Figure 13. The body thickness distribution of the subject Obj may be a measured value or an estimated value, so the body thickness distribution acquisition unit 61 acquires either a measured value of the body thickness distribution of the subject Obj or an estimated value of the body thickness distribution of the subject Obj. Based on the acquired body thickness distribution, the body thickness distribution acquisition unit 61 estimates scattered radiation for each pixel of the image to be corrected and removes scattered radiation from each pixel of the image to be corrected. The image to be corrected after the removal of scattered radiation is completed proceeds to the next processing step.

[0127] In this embodiment, the images to be corrected are all images of a specific subject, and the first radiation image G1 and the second radiation image G2 are radiation images obtained almost simultaneously by the so-called one-shot ES method. Therefore, the thickness distribution of the subject's obj can be estimated based on either of the images to be corrected, and the obtained estimate can be used as the thickness distribution for all the images to be corrected.

[0128] As shown in Figure 14, the body thickness distribution acquisition unit 61 includes a body thickness distribution measurement value acquisition unit 62, a body thickness distribution estimated value acquisition unit 63, and a scattered radiation removal unit 64. The body thickness distribution measurement value acquisition unit 62 acquires the measured body thickness distribution. The body thickness distribution measurement value acquisition unit 62 acquires the measurement value obtained by actually measuring the body thickness distribution and uses this as the body thickness distribution of the subject Obj.

[0129] The body thickness distribution measurement unit 62 calculates the body thickness distribution T(x, y) of the subject H based on the SID (Source Image receptor Distance) and SOD (Source Object Distance) included in the imaging conditions. 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, laser rangefinder, etc.

[0130] The body thickness distribution is preferably determined by subtracting the SOD from the SID. The body thickness distribution is calculated pixel by pixel, corresponding to the first and second radiation images G1 and G2. Alternatively, instead of calculating the body thickness distribution based on the SID and SOD, the body thickness distribution may be calculated from at least one of the first and second radiation images G1 and G2. Alternatively, the body thickness distribution may be calculated from the soft tissue image of the subject H obtained by performing weighted subtraction between the corresponding pixels of the first radiation image G1 and the second radiation image G2. When determining the body thickness distribution, if the first and second radiation detectors 44 and 45 are located inside a stage (not shown) on which the subject Obj is placed, it is preferable to use the distance between the radiation source 41 and the surface of the stage in contact with the subject Obj as the SID.

[0131] The body thickness distribution estimation unit 63 obtains an estimated value of the body thickness distribution of the subject using a known method. As a method for obtaining the estimated value of the body thickness distribution, or as a method for performing image processing for scattering removal using the obtained estimated value of the body thickness distribution, in addition to the method using the LUT (Figure 5) described above, a method using a virtual model described in Japanese Patent Application Publication No. 2015-043959 can also be adopted.

[0132] The method using a virtual model first acquires a virtual model with a predetermined body thickness distribution, and then generates an estimated image by combining the estimated primary radiation image and estimated scattered radiation image obtained from radiography of the virtual model. Next, the body thickness distribution of the virtual model is modified from the subject image obtained from radiography of the subject Obj and the estimated image so that the error value representing the difference in pixel values ​​at each corresponding position is reduced. The modified thickness distribution of the virtual model is determined as the body thickness distribution of the subject. In this way, even for subjects photographed without using a grid, for example, the influence of scattered radiation can be suppressed and a more accurate estimate of the body thickness distribution can be obtained. The acquired body pressure distribution is used for contrast correction processing, scattered radiation correction processing, etc.

[0133] Preferably, the scattered radiation removal unit 64 uses the acquired body pressure distribution to remove scattered radiation from the first radiation image G1 or the second radiation image G2 by a scattered radiation correction process, and then performs image processing such as bone extraction or soft tissue extraction to generate a bone image Gb or a soft tissue image Gt. Because the scattered radiation correction process suppresses the effects of scattered radiation and allows various image processing to be performed, it is possible to generate a bone image Gb or a soft tissue image Gt with better image quality.

[0134] The display control unit 14 controls the display of the first enhanced image, level value, and image quality index on the display unit 16. It is preferable to set a specific range in the image quality index as a reference range and to control the display of the image quality index and the reference range on the display unit 16. The specific range to be set as the reference range can be set in advance. For example, the reference range can be the range of the ES image quality index in which, if the ES image quality index is within this reference range, the image quality of the ES image is sufficiently good and there is no need to perform any further image quality adjustments. Furthermore, by setting a reference range, it is possible to grasp at a glance whether the ES image quality of the ES image is within a certain standard or not.

[0135] As shown in Figure 15, in this embodiment, the reference range on the number line 31 is indicated by a rectangle 33 as a specific range centered on the 0 mark. If the point 32 indicating the ES quality index of the bone image Gb displayed on screen 21 is not inside the rectangle 33 indicating the reference range, then the ES quality of the bone image Gb displayed on screen 21 is outside the reference range, and it is necessary to adjust the ES quality to improve the ES quality index. Therefore, it is immediately clear whether or not adjustments to the ES quality, etc., are necessary for the bone image Gb displayed on screen 21.

[0136] As shown in Figure 15, if the image quality index of the generated first-weighted image is outside the reference range, the user can generate a first-weighted image with improved ES quality by repeatedly changing the ES level value. Also, as shown in Figure 16, if the image quality index of the generated first-weighted image is within the reference range, it is immediately clear that no image quality adjustment is necessary.

[0137] As described above, by setting a specific range in the image quality index as a reference range and displaying it on the screen, it is possible to instantly see whether the image quality of each ES image is above a certain standard. Furthermore, even if the image quality does not meet the standard, it is possible to simply change the ES level value multiple times, and the changed ES image quality index will be displayed each time, so it is possible to understand the trend of how much the ES image quality index improves or deteriorates depending on how the values ​​are changed. Therefore, it is easy and quick to adjust the ES image quality to be within the reference range.

[0138] Furthermore, the processing unit 10 may notify the user if the image quality index is not within the reference range. In this case, as shown in Figure 17, the processing unit 10 includes a notification control unit 71. The notification only needs to be in a manner that allows the user to understand that the image quality index is not within the reference range, and may be in any form, such as notification by display on a display, notification by sound, and / or notification by vibration of a mobile device communicating with the processing unit 10.

[0139] In this embodiment, notification is provided by displaying information on the display unit 16. As shown in Figure 18, for example, when displaying the ES image quality index, if the image quality index for the generated first weighted image is outside the reference range, a warning display 72 can be displayed or flashed on the ES image quality index display unit 24. Alternatively, the background color of the entire ES image quality index display unit 24 may be displayed in a different color than usual. This allows the user, a radiologic technologist, to recognize that further adjustments are needed for the ES image currently displayed on the display, even if they are not paying particular attention to the display unit 16.

[0140] As described above, if the image quality index is not within the reference range, the radiographer will be notified, allowing them to recognize that the ES image quality index is not within the reference range and that adjustment is necessary, even if they are not paying particular attention to the display unit 16. Therefore, this is particularly effective when the user has to generate a large number of ES images.

[0141] Furthermore, although the above embodiment used a radiographic image of a human chest as the specific subject, any ES image can be used as the specific subject, and it can also be applied to other parts of a person or animal, or tissues such as blood vessels. In addition, since ES images can be used to observe changes over time, it can also be applied to ES images taken to observe changes over time. Therefore, the specific subject can be not only a person or an animal, but also a building, such as a crack in a pipe or structure.

[0142] Furthermore, in the above embodiment, the first weighted image or the second weighted image is an image generated based on the acquired radiation image, but it may also be an image whose image quality has been adjusted by the processing device 10. For example, when generating the first weighted image, the second weighted image whose image quality has been adjusted by the processing device 10 may be used. This makes it possible to obtain an ES image with improved image quality.

[0143] In the above embodiment, the hardware structure of the processing units, such as the image acquisition unit 11, image processing unit 12, correlation information generation unit 13, display control unit 14, change reception unit 15, notification control unit 71, or central control unit (not shown) included in the processing device 10, or the central control unit (not shown) included in the console 43, is one of the following types of processors. These types of processors include a CPU (Central Processing Unit), which is a general-purpose processor that executes software (programs) and functions as various processing units; a 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 electrical circuit, which is a processor with a circuit configuration specifically designed to execute various processes.

[0144] 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 clients and servers. 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 a System on a Chip (SoC). Thus, various processing units are configured, in terms of hardware structure, using one or more of the above-mentioned various processors.

[0145] Furthermore, the hardware structure of these various processors is, more specifically, an electrical circuit in the form of a combination of circuit elements such as semiconductor devices.

[0146] From the above description, the radiation image processing equipment described in appendices 1 to 16 below can be understood.

[0147] [Note 1] Equipped with a processor, The processor is Two radiation images are obtained by using two different types of radiation energy to photograph a specific subject. A first enhanced image is generated by performing a first enhancement process on the two aforementioned radiation images using a calculation formula that includes a first parameter. A second-weighted image is generated by performing a second weighting process using at least one of the two aforementioned radiation images. Correlation information showing the correlation between the first-weighted image and the second-weighted image is generated. Control is performed to display the first enhanced image, a level value corresponding to the value of the first parameter, and an image quality index that quantifies the degree of difference between the first enhanced image and the second enhanced image based on the correlation information, on the display unit. A radiation image processing device that accepts changes to the aforementioned level value.

[0148] [Note 2] When the processor receives a change in the level value, it updates the value of the first parameter. Using the updated value of the first parameter, the first enhanced image is generated. The radiation image processing apparatus according to Appendix 1, which displays the updated level value corresponding to the updated value of the first parameter on the display unit.

[0149] [Note 3] The radiographic image processing apparatus according to Appendix 1 or 2, wherein the processor controls the display unit to display a user interface that accepts changes to the level value by the user.

[0150] [Note 4] The radiation image processing apparatus according to any one of appendices 1 to 3, wherein the radiation image is obtained by removing the scattered radiation component estimated according to the thickness of the subject for each pixel.

[0151] [Note 5] The first enhancement process is a subtraction process, as described in any one of appendices 1 to 4, for the radiation image processing apparatus.

[0152] [Note 6] The subject includes both bony and soft tissue. The radiographic image processing apparatus according to any one of appendices 1 to 5, wherein the first weighted image is a bone image and the second weighted image is a soft tissue image.

[0153] [Note 7] The subject includes both bony and soft tissue. The radiographic image processing apparatus according to any one of appendices 1 to 5, wherein the first weighted image is a soft tissue image and the second weighted image is a bone image.

[0154] [Note 8] The radiation image processing apparatus according to any one of appendices 1 to 7, wherein the first enhancement process includes a process of weighting one of the two radiation images using the first parameter and then subtracting it from the other radiation image.

[0155] [Note 9] The radiation image processing apparatus according to any one of appendices 1 to 8, wherein the second enhancement process includes a process of weighting one of the two radiation images using the second parameter and then subtracting it from the other radiation image.

[0156] [Note 10] The subject includes bone and soft tissue, The first weighted image is a bone image, and the second weighted image is a soft tissue image. If, among the two aforementioned radiographic images, the pixel value at coordinate (x,y) in one radiographic image is G1(x,y), the pixel value at coordinate (x,y) in the other radiographic image is G2(x,y), the pixel value at coordinate (x,y) in the bone image is Gb(x,y), and the pixel value at coordinate (x,y) in the soft tissue image is Gt(x,y), and the first parameter is α and the second parameter is β, The radiographic image processing apparatus according to any one of Appendix 1 to 5, wherein the bone image is generated by the following formula (1), and the soft tissue image is generated by the following formula (4).

[0157] Gb(x,y)=G1(x,y)-α×G2(x,y) (1) Gt(x,y)=G1(x,y)-β×G2(x,y) (4)

[0158] [Note 11] The radiation image processing apparatus according to any one of appendices 1 to 8, wherein the second enhancement process is a process of subtracting the first enhanced image from one of the two radiation images.

[0159] [Note 12] The subject includes bone and soft tissue, The first weighted image is a bone image, and the second weighted image is a soft tissue image. If, of the two aforementioned radiation images, the pixel value at coordinate (x,y) in one radiation image is G1(x,y), and the pixel value at coordinate (x,y) in the other radiation image is G2(x,y), and the pixel value at coordinate (x,y) in the bone image is Gb(x,y), the pixel value at coordinate (x,y) in the soft tissue image is Gt(x,y), and the first parameter is α, The radiographic image processing apparatus according to any one of Appendix 1 to 5, wherein the bone image is generated by the following formula (1), and the soft tissue image is generated by the following formula (2) or formula (3).

[0160] Gb(x,y)=G1(x,y)-α×G2(x,y) (1) Gt(x,y)=G1(x,y)-Gb(x,y) (2) Gt(x,y) = G2(x,y) - Gb(x,y) (3)

[0161] [Note 13] The correlation information is the correlation coefficient between the pixel values ​​of the first-enhanced image and the pixel values ​​of the second-enhanced image, as described in any one of Appendix 1 to 12.

[0162] [Note 14] The image quality index is the correlation coefficient, as described in Appendix 13, for the radiation image processing apparatus.

[0163] [Note 15] The radiographic image processing apparatus according to any one of appendices 1 to 14, wherein the processor sets a specific range in the image quality index as a reference range and controls the display unit to display the image quality index and the reference range.

[0164] [Note 16] The radiographic image processing apparatus described in Appendix 15, wherein the processor performs control to notify the user if the image quality index is not included in the reference range. [Explanation of Symbols]

[0165] 10. Radiation image processing device 11 Image acquisition unit 12 Image Processing Unit 13. Correlation Information Generation Unit 14 Display Control Unit 15 Change Request Department 16 Display 17 Control section 20 LUT 21 screens 22 Image Adjustment Section 23 ES Image Switching Section 24 ES Image Quality Index Display Unit 25 ES level display section 31 Number Line 32 points 33 rectangle 34 ES level value display section 35 ES Level Value Change Section 36 Cursors 40. Radiography System 41 Radiation source 42. Radiography Panel 43 Console 44. First radiation detector 45. Second Radiation Detector 46 Radiation energy conversion filter 47 Display section 48 Control section 51 RIS 52 HIS 53 PACS 61 Body thickness distribution acquisition part 62 Body thickness distribution measurement acquisition unit 63 Body Thickness Distribution Estimation Unit 64 Scattered radiation removal section 71 Notification Control Unit 72 Warning display G1 First Radiation Image G2 Second Radiation Image Gb bone image Gt Soft tissue images Obj Subject Ra X-ray ST110~ST180 Step

Claims

1. Equipped with a processor, The processor is Two radiation images are obtained by using two different types of radiation energy to photograph a specific subject. A first enhanced image is generated by performing a first enhancement process on the two aforementioned radiation images using a calculation formula that includes a first parameter. A second-enhanced image is generated by performing a second enhancement process using at least one of the two aforementioned radiation images. Correlation information showing the correlation between the first-enhanced image and the second-enhanced image is generated. Control is performed to display the first enhanced image, a level value corresponding to the value of the first parameter, and an image quality index that quantifies the degree of difference between the first enhanced image and the second enhanced image based on the correlation information, on the display unit. Accepting changes to the aforementioned level value, The level value is obtained by transforming the first parameter so that the relationship between the value of the first parameter and the image quality based on the correlation information can be recognized.

2. When the processor receives a change in the level value via the user interface, and improves the image quality index by changing the first parameter according to the level value, it updates the value of the changed first parameter. Using the updated value of the first parameter, the first enhanced image is generated. The radiation image processing apparatus according to claim 1, which displays the updated level value corresponding to the updated value of the first parameter on the display unit.

3. The radiation image processing apparatus according to claim 1 or 2, wherein the processor controls the display unit to display a user interface that accepts changes to the level value by the user.

4. The radiation image processing apparatus according to claim 1 or 2, wherein the radiation image is obtained by removing the scattered radiation component estimated according to the thickness of the subject for each pixel.

5. The radiation image processing apparatus according to claim 1 or 2, wherein the first enhancement process is a subtraction process.

6. The subject includes bone and soft tissue, The radiographic image processing apparatus according to claim 1 or 2, wherein the first weighted image is a bone image and the second weighted image is a soft tissue image.

7. The subject includes bone and soft tissue, The radiographic image processing apparatus according to claim 1 or 2, wherein the first weighted image is a soft tissue image and the second weighted image is a bone image.

8. The radiation image processing apparatus according to claim 1 or 2, wherein the first enhancement process includes a process of weighting one of the two radiation images using the first parameter and then subtracting it from the other radiation image.

9. The radiation image processing apparatus according to claim 1 or 2, wherein the second enhancement process is performed by an arithmetic formula including a second parameter, and the process includes weighting one of the two radiation images using the second parameter and then subtracting it from the other radiation image.

10. The subject includes bone and soft tissue, The first weighted image is a bone image, and the second weighted image is a soft tissue image. If, among the two aforementioned radiographic images, the pixel value at coordinate (x,y) in one radiographic image is G1(x,y), the pixel value at coordinate (x,y) in the other radiographic image is G2(x,y), the pixel value at coordinate (x,y) in the bone image is Gb(x,y), and the pixel value at coordinate (x,y) in the soft tissue image is Gt(x,y), the first parameter is α, and the second parameter is β, The radiographic image processing apparatus according to claim 9, wherein the bone image is generated by the following formula (1), and the soft tissue image is generated by the following formula (4). Gb(x,y)=G1(x,y)−α×G2(x,y) (1) Gt(x,y)=G1(x,y)−β×G2(x,y) (4)

11. The radiation image processing apparatus according to claim 1 or 2, wherein the second enhancement process is a process of subtracting the first enhancement image from one of the two radiation images.

12. The subject includes bone and soft tissue, The first weighted image is a bone image, and the second weighted image is a soft tissue image. If, among the two aforementioned radiation images, the pixel value at coordinate (x,y) in one radiation image is G1(x,y), the pixel value at coordinate (x,y) in the other radiation image is G2(x,y), the pixel value at coordinate (x,y) in the bone image is Gb(x,y), the pixel value at coordinate (x,y) in the soft tissue image is Gt(x,y), and the first parameter is α, The radiographic image processing apparatus according to claim 1 or 2, wherein the bone image is generated by the following formula (1), and the soft tissue image is generated by the following formula (2) or formula (3). Gb(x,y)=G1(x,y)−α×G2(x,y) (1) Gt (x, y) = G1 (x, y) - Gb (x, y) (2) Gt (x, y) = G2 (x, y) - Gb (x, y) (3)

13. The radiation image processing apparatus according to claim 1 or 2, wherein the correlation information is the correlation coefficient between the pixel values ​​of the first enhanced image and the pixel values ​​of the second enhanced image.

14. The radiation image processing apparatus according to claim 13, wherein the image quality index is the correlation coefficient.

15. The radiation image processing apparatus according to claim 1 or 2, wherein the processor sets a specific range in the image quality index as a reference range and controls the display unit to display the image quality index and the reference range.

16. The radiation image processing apparatus according to claim 15, wherein the processor performs control to notify the user if the image quality index is not included in the reference range.

17. The steps include: obtaining two radiation images of a specific subject using two different types of radiation energy, The steps include generating a first enhanced image by performing a first enhancement process on the two aforementioned radiation images using a calculation formula that includes a first parameter, A step of generating a second-enhanced image by performing a second enhancement process using at least one of the two aforementioned radiation images, A step of generating correlation information showing the correlation between the first-enhanced image and the second-enhanced image, The system includes a step of controlling the display unit to display the first enhanced image, a level value corresponding to the value of the first parameter, and an image quality index that quantifies the degree of difference between the first enhanced image and the second enhanced image based on the correlation information. A method for operating a radiation image processing device, which is obtained by converting the first parameter so that the level value can recognize the relationship between the value of the first parameter and the image quality based on the correlation information.

18. A function to acquire two radiation images of a specific subject using two different types of radiation energy, A function to generate a first enhanced image by performing a first enhancement process on the two aforementioned radiation images using a calculation formula that includes a first parameter, A function to generate a second-enhanced image by performing a second enhancement process using at least one of the two aforementioned radiation images, A function to generate correlation information showing the correlation between the first-enhanced image and the second-enhanced image, The computer is instructed to perform a function that controls the display unit to display the first enhanced image, a level value corresponding to the value of the first parameter, and an image quality index that quantifies the degree of difference between the first enhanced image and the second enhanced image based on the correlation information. The level value is a radiation image processing program obtained by converting the first parameter so that the relationship between the value of the first parameter and the image quality based on the correlation information can be recognized.

Images