WO2022181022A1 - Image processing device and method, radiography system, and program - Google Patents

Abstract

This image processing device uses a plurality of moving images corresponding to a plurality of different types of energy to perform energy subtraction processing and thereby generate moving images of separated images, and corrects the plurality of moving images or the moving images of the separated images so as to reduce fluctuations in the signal value between frames in the plurality of moving images or the moving images of the separated images.

Description

IMAGE PROCESSING APPARATUS AND METHOD, RADIATION IMAGING SYSTEM, PROGRAM

The present invention relates to an image processing apparatus and method, a radiation imaging system, and a program.

Currently, radiographic imaging devices that use flat panel detectors (Flat Panel Detectors, hereinafter abbreviated as FPDs) made of semiconductor materials are widely used as imaging devices for X-ray medical image diagnosis and non-destructive inspection. Such radiation imaging apparatuses are used, for example, in medical image diagnosis as digital imaging apparatuses for still image capturing such as general radiography and moving image capturing such as fluoroscopic imaging. One of imaging methods using an FPD is energy subtraction (Patent Document 1). In energy subtraction, a plurality of images corresponding to X-rays of different energies are acquired, and an image of a specific material (for example, a bone image and a soft tissue image) is obtained from the plurality of images by utilizing the difference in the X-ray attenuation rate of the material. tissue images) are separated.

Patent Document 2 describes a system that performs dynamic dual-energy imaging by applying energy subtraction processing to moving images. In this system, during imaging, the tube voltage of the X-ray source is changed to the first kV value and then to the second kV value. Then, the first signal corresponding to the first sub-image is integrated when the tube voltage is at the first kV value, and the integration is reset after the integrated signal is transferred to the sample and hold node. A second signal corresponding to a second sub-image is then integrated when the tube voltage is at a second kV value. As a result, readout of the integrated first signal and integration of the second signal are performed in parallel. According to the system of Patent Literature 2, an energy subtraction moving image can be captured.

JP 2019-162358 A Special Table 2009-504221

However, the above-described conventional technology has a problem that flickering occurs in the moving image because the pixel values of the energy subtraction moving image vary between frames due to temporal output fluctuations between sub-image frames. Energy-subtracted movies with flickering may lead to inaccurate diagnostic imaging.

The present invention provides a technique for reducing flickering in energy subtraction moving images.

An image processing apparatus according to one aspect of the present invention has the following configuration. i.e.
generating means for generating moving images of separated images by performing energy subtraction processing using a plurality of moving images corresponding to a plurality of different radiation energies;
and correction means for correcting the moving images of the plurality of moving images or the moving images of the separated images so as to reduce variations in signal values between frames in the moving images of the plurality of moving images or the moving images of the separated images.

According to the present invention, flickering in energy subtraction moving images is reduced.

Other features and advantages of the present invention will become apparent from the following description with reference to the accompanying drawings. In the accompanying drawings, the same or similar configurations are given the same reference numerals.

The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.
1 is a diagram showing a configuration example of a radiation imaging system according to an embodiment; FIG. FIG. 2 is an equivalent circuit diagram of pixels included in a two-dimensional detector of the X-ray imaging apparatus; 4 is a timing chart showing operations for acquiring an X-ray image; 4 is a timing chart showing operations for acquiring an X-ray image; FIG. 4 is a block diagram of correction processing in energy subtraction processing; A block diagram of signal processing in energy subtraction processing. FIG. 3 is a block diagram of image processing related to energy subtraction processing; FIG. 2 is a block diagram showing a hardware configuration example of a control computer; FIG. 4 is a diagram showing an example of accumulated images and bone images; FIG. 4 is a diagram showing signal value fluctuations between frames of a moving image before and after energy subtraction processing; 4 is a block diagram of signal processing including blink reduction processing according to the first embodiment; FIG. 4 is a flowchart of flickering reduction processing according to the embodiment; FIG. 4 is a diagram showing signal value fluctuations between frames of a moving image subjected to flickering reduction processing; 4A and 4B are diagrams for explaining a process of acquiring an accumulated image A; FIG. FIG. 11 is a block diagram of signal processing including flicker reduction processing and noise reduction processing according to the second embodiment;

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. In addition, the following embodiments do not limit the invention according to the scope of claims. Although multiple features are described in the embodiments, not all of these multiple features are essential to the invention, and multiple features may be combined arbitrarily. Furthermore, in the accompanying drawings, the same or similar configurations are denoted by the same reference numerals, and redundant description is omitted.

Although a radiation imaging system using X-rays as radiation will be described below, it is not limited to this. Radiation in the present invention includes alpha rays, beta rays, and gamma rays, which are beams produced by particles (including photons) emitted by radioactive decay, as well as beams having radiation energy at the same level or higher, such as particle beams. , cosmic rays, etc. are also included.

(First embodiment)
FIG. 1 is a block diagram showing a configuration example of a radiation imaging system according to the first embodiment. The radiation imaging system of the first embodiment includes an X-ray generation device 101 , an X-ray control device 102 , a control computer 103 and an X-ray imaging device 104 .

The X-ray generator 101 emits X-rays. The X-ray controller 102 controls X-ray irradiation by the X-ray generator 101 . The control computer 103 controls the X-ray imaging device 104 to acquire a radiographic image (hereinafter referred to as an X-ray image (image information)) captured by the X-ray imaging device 104 . The control computer 103 functions as an image processing device that performs image processing including flickering reduction processing, which will be described later, on an X-ray image acquired from the X-ray imaging device 104 . Note that the X-ray imaging apparatus 104 may be provided with a function of executing image processing including flickering reduction processing (and noise reduction processing described in the second embodiment). The X-ray imaging device 104 is composed of a phosphor 105 that converts X-rays into visible light and a two-dimensional detector 106 that detects visible light. The two-dimensional detector 106 is a sensor in which pixels 20 for detecting X-ray quanta are arranged in an array of X columns×Y rows, and outputs image information.

FIG. 5 is a block diagram showing a hardware configuration example of the control computer 103. As shown in FIG. The CPU 141 controls various operations of the control computer 103 by executing programs stored in the ROM 142 or RAM 143 . For example, the CPU 141 controls X-ray irradiation by the X-ray control device 102 (X-ray generator 101 ) and X-ray image capturing operation by the X-ray imaging device 104 . The CPU 141 also implements various signal processing and image processing, which will be described later. It should be noted that the operation of signal processing and image processing, which will be described later, may be partially or wholly realized by dedicated hardware. The ROM 142 stores programs executed by the CPU 141 and various data. The RAM 143 provides a work area for storing intermediate data generated when the CPU 141 executes processing. The secondary storage device 144 stores radiation images (X-ray images) to be processed. The secondary storage device 144 also stores control programs. The programs stored in the secondary storage device 144 are developed in the RAM 143 as necessary and executed by the CPU 141 .

The display 145 performs various displays under the control of the CPU 141. The operating unit 146 includes, for example, a keyboard and a pointing device, and receives various user inputs. The interface 147 connects external devices such as the X-ray control device 102 and the X-ray imaging device 104 to the control computer 103 . A bus 148 communicably connects the units described above.

FIG. 2 is an equivalent circuit diagram of the pixel 20 included in the two-dimensional detector 106. FIG. The pixel 20 includes a photoelectric conversion element 201 and an output circuit section 202 . Photoelectric conversion element 201 can typically be a photodiode. The output circuit section 202 includes an amplifier circuit section 204 , a clamp circuit section 206 , a sample hold circuit section 207 and a selection circuit section 208 .

The photoelectric conversion element 201 includes a charge storage section, and the charge storage section is connected to the gate of the MOS transistor 204 a of the amplifier circuit section 204 . The source of MOS transistor 204a is connected to current source 204c through MOS transistor 204b. A source follower circuit is formed by the MOS transistor 204a and the current source 204c. The MOS transistor 204b is an enable switch that turns on when the enable signal EN supplied to its gate becomes active level to put the source follower circuit into operation.

In the example shown in FIG. 2, the charge accumulating portion of the photoelectric conversion element 201 and the gate of the MOS transistor 204a constitute a common node, and this node converts the charge accumulated in the charge accumulating portion into a voltage. It functions as a voltage converter. That is, a voltage V (=Q/C) appears in the charge-voltage converter, which is determined by the charge Q accumulated in the charge storage and the capacitance value C of the charge-voltage converter. The charge-voltage converter is connected to reset potential Vres through reset switch 203 . When the reset signal PRES becomes active level, the reset switch 203 is turned on, and the potential of the charge-voltage converter is reset to the reset potential Vres.

The clamp circuit section 206 clamps the noise output by the amplifier circuit section 204 according to the reset potential of the charge-voltage conversion section with the clamp capacitor 206a. In other words, the clamp circuit unit 206 is a circuit for canceling this noise from the signal output from the source follower circuit according to the charge generated by photoelectric conversion in the photoelectric conversion element 201 . This noise includes kTC noise at reset. Clamping is performed by setting the clamp signal PCL to the active level to turn on the MOS transistor 206b and then setting the clamp signal PCL to the inactive level to turn off the MOS transistor 206b. The output side of the clamp capacitor 206a is connected to the gate of the MOS transistor 206c. The source of MOS transistor 206c is connected to current source 206e through MOS transistor 206d. A source follower circuit is formed by the MOS transistor 206c and the current source 206e. The MOS transistor 206d is an enable switch that turns on when the enable signal EN0 supplied to its gate becomes active level to put the source follower circuit into operation.

A signal output from the clamp circuit unit 206 according to the charge generated by photoelectric conversion in the photoelectric conversion element 201 is written as a light signal into the capacitor 207Sb via the switch 207Sa when the light signal sampling signal TS becomes active level. be The signal output from the clamp circuit section 206 when the MOS transistor 206b is turned on immediately after resetting the potential of the charge-voltage conversion section is the clamp voltage. This noise signal is written into the capacitor 207Nb through the switch 207Na when the noise sampling signal TN becomes active level. This noise signal contains the offset component of the clamp circuit section 206 . A switch 207Sa and a capacitor 207Sb constitute a signal sample and hold circuit 207S, and a switch 207Na and a capacitor 207Nb constitute a noise sample and hold circuit 207N. The sample and hold circuit section 207 includes a signal sample and hold circuit 207S and a noise sample and hold circuit 207N.

When the drive circuit drives the row selection signal to the active level, the signal (light signal) held in the capacitor 207Sb is output to the signal line 21S via the MOS transistor 208Sa and the row selection switch 208Sb. At the same time, the signal (noise) held in capacitor 207Nb is output to signal line 21N via MOS transistor 208Na and row select switch 208Nb. The MOS transistor 208Sa forms a source follower circuit with a constant current source (not shown) provided on the signal line 21S. Similarly, the MOS transistor 208Na forms a source follower circuit with a constant current source (not shown) provided on the signal line 21N. A signal selection circuit portion 208S is composed of the MOS transistor 208Sa and the row selection switch 208Sb, and a noise selection circuit portion 208N is composed of the MOS transistor 208Na and the row selection switch 208Nb. The selection circuit section 208 includes a signal selection circuit section 208S and a noise selection circuit section 208N.

The pixel 20 may have an addition switch 209S that adds the optical signals of a plurality of adjacent pixels 20. In the addition mode, the addition mode signal ADD becomes active level and the addition switch 209S is turned on. As a result, the capacitors 207Sb of adjacent pixels 20 are connected to each other by the addition switch 209S, and the optical signals are averaged. Similarly, pixel 20 may have a summing switch 209N that sums the noise of adjacent pixels 20 . When the adder switch 209N is turned on, the capacitors 207Nb of the adjacent pixels 20 are interconnected by the adder switch 209N to average noise. Addition section 209 includes an addition switch 209S and an addition switch 209N.

The pixel 20 may have a sensitivity changing section 205 for changing sensitivity. The pixel 20 can include, for example, a first sensitivity change switch 205a and a second sensitivity change switch 205'a and their associated circuit elements. When the first change signal WIDE becomes active level, the first sensitivity change switch 205a is turned on, and the capacitance value of the first additional capacitor 205b is added to the capacitance value of the charge-voltage converter. This reduces the sensitivity of the pixel 20 . When the second change signal WIDE2 becomes active level, the second sensitivity change switch 205'a is turned on, and the capacitance value of the second additional capacitor 205'b is added to the capacitance value of the charge-voltage converter. This further reduces the sensitivity of the pixel 20 . By adding the function of lowering the sensitivity of the pixel 20 in this way, it becomes possible to receive a larger amount of light, and the dynamic range can be widened. When the first change signal WIDE becomes the active level, the enable signal ENw may be made the active level to cause the MOS transistor 204'a to perform the source follower operation instead of the MOS transistor 204a.

The X-ray imaging apparatus 104 reads the output of the pixel circuit as described above, converts it into a digital value with an AD converter (not shown), and then transfers the image to the control computer 103 .

Next, the operation of the radiation imaging system of this embodiment (driving of the X-ray imaging device 104) will be described. 3A and 3B are diagrams showing drive timings of the X-ray imaging apparatus 104 according to the first embodiment. In FIG. 3A , the horizontal axis represents time, and the timings of X-ray irradiation, synchronization signals, resetting of the photoelectric conversion element 201 , sample hold circuit 207 and image reading from the signal line 21 are shown.

First, the photoelectric conversion element 201 is reset, and then X-rays are emitted. Although the X-ray tube voltage ideally becomes a rectangular wave, it takes a finite amount of time for the tube voltage to rise and fall. In particular, when the exposure time is short with pulsed X-rays, the tube voltage can no longer be regarded as a rectangular wave, and has a waveform as shown in FIG. 3A. That is, the radiation energy (energy of X-rays) differs in the rising period, the stable period, and the falling period of X-rays.

Therefore, sampling is performed by the noise sample-and-hold circuit 207N after the X-ray 301 in the rising period is emitted, and sampling is performed by the signal sample-and-hold circuit 207S after the X-ray 302 in the stable period is emitted. After that, the difference between the signal lines 21N and 21S is read out as an image. At this time, the noise sample-and-hold circuit 207N holds the signal (R ₁ ) of the X-rays 301 in the rising period, and the signal sample-and-hold circuit 207S holds the signal of the X-rays 301 in the rising period and the signal of the X-rays 302 in the stable period. (R ₁ +B) is retained. Therefore, an image 304 corresponding to the signal (B) of the X-rays 302 in the stable period is read out from the X-ray imaging apparatus 104 .

Next, after the exposure of the X-rays 303 in the fall period and the reading of the image 304 are completed, sampling is performed again by the signal sample-and-hold circuit 207S. Thereafter, the photoelectric conversion element 201 is reset, sampling is performed again by the noise sample hold circuit 207N, and the difference between the signal lines 21N and 21S is read out as an image. At this time, the noise sample-and-hold circuit 207N holds the signal in the state where the X-ray is not irradiated. The signal sample-and-hold circuit 207S holds the sum (R ₁ +B+R ₂ ) of the signal of the X-ray 301 in the rising period, the X-ray 302 in the stable period, and the signal of the X-ray 303 in the falling period. Therefore, from the X-ray imaging apparatus 104, an image 306 corresponding to the signal of the X-rays 301 in the rise period, the signal of the X-rays 302 in the stable period, and the signal of the X-rays 303 in the fall period is read. After that, by calculating the difference between the image 306 and the image 304, an image 305 corresponding to the sum (R ₁ +R ₂ ) of the X-rays 301 in the rising period and the X-rays 303 in the falling period is obtained.

The timing for resetting the sample-and-hold circuit 207 and the photoelectric conversion element 201 is determined using the synchronization signal 307 indicating that the X-ray generator 101 has started to emit X-rays. As a method for detecting the start of X-ray irradiation, a configuration that measures the tube current of the X-ray generator 101 and determines whether or not the current value exceeds a preset threshold value is preferably used. Also, after the photoelectric conversion element 201 is completely reset, the pixel 20 is repeatedly read out, and a configuration in which it is determined whether or not the pixel value exceeds a preset threshold value is preferably used. Furthermore, a configuration in which an X-ray detector different from the two-dimensional detector 106 is incorporated in the X-ray imaging apparatus 104 and whether or not the measured value exceeds a preset threshold is preferably used. In either method, after a predetermined time has passed from the input of the synchronous signal 307, sampling of the signal sample and hold circuit 207S, sampling of the noise sample and hold circuit 207N, and resetting of the photoelectric conversion element 201 are performed.

As described above, an image 304 corresponding to the stable period of the pulse X-ray and an image 305 corresponding to the sum of the rising period and the falling period are obtained. Since the energies of the X-rays irradiated when forming the two images are different, energy subtraction processing can be performed by performing calculations between the images.

FIG. 3B shows drive timing when energy subtraction is performed in the radiation imaging system according to the first embodiment. This differs from FIG. 3A in that the X-ray tube voltage is actively switched.

First, the photoelectric conversion element 201 is reset, and then low-energy X-rays 401 are irradiated. After that, sampling is performed by the noise sample-and-hold circuit 207N, and after the tube voltage is switched and the high-energy X-ray 402 is emitted, sampling is performed by the signal sample-and-hold circuit 207S. Thereafter, the tube voltage is switched to irradiate low-energy X-rays 403 . Furthermore, the difference between the signal lines 21N and 21S is read out as an image. At this time, the noise sample-and-hold circuit 207N holds the signal (R ₁ ) of the low-energy X-rays 401, and the signal sample-and-hold circuit 207S holds the signal of the low-energy X-rays 401 and the signal of the high-energy X-rays 402. (R ₁ +B) is retained. Therefore, an image 404 corresponding to the signal (B) of high-energy X-rays 402 is read out from the X-ray imaging apparatus 104 .

Next, after the irradiation of the low-energy X-rays 403 and the reading of the image 404 are completed, sampling is performed again by the signal sample-and-hold circuit 207S. Thereafter, the photoelectric conversion element 201 is reset, sampling is performed again by the noise sample hold circuit 207N, and the difference between the signal lines 21N and 21S is read out as an image. At this time, the noise sample-and-hold circuit 207N holds the signal in the state where the X-ray is not irradiated. At this time, the sum of the signals of the low-energy X-ray 401, the high-energy X-ray 402, and the low-energy X-ray 403 (R ₁ +B+R ₂ ) is held in the signal sample-and-hold circuit 207S. Therefore, an image 406 corresponding to the low-energy X-ray 401 signal, the high-energy X-ray 402 signal, and the low-energy X-ray 403 signal is read out from the X-ray imaging apparatus 104 . After that, by calculating the difference between the image 406 and the image 404, an image 405 corresponding to the sum (R ₁ +R ₂ ) of the low energy X-ray 401 and the low energy X-ray 403 is obtained. The sync signal 407 is the same as in FIG. 3A. By acquiring images while actively switching the tube voltage in this way, the energy difference between the low-energy and high-energy images can be made larger than in the method of FIG. 3A.

Next, the energy subtraction processing of this embodiment will be described with reference to FIGS. 4A to 4C. The energy subtraction processing in this embodiment is divided into three stages of correction processing, signal processing, and image processing. Processing at each stage will be described below.

Description of Correction Processing FIG. 4A is a block diagram of correction processing in the energy subtraction processing according to this embodiment. First, imaging is performed without exposing the X-ray imaging device 104 to X-rays, and an image is acquired by the driving shown in FIG. 3A or 3B. At this time, two images are read, the first image being F_ODD and the second image being F_EVEN. F_ODD and F_EVEN are images corresponding to fixed pattern noise (FPN) of the X-ray imaging device 104 . Next, X-rays are emitted to the X-ray imaging device 104 in the absence of a subject to perform imaging, and an image is acquired by the driving shown in FIG. 3A or 3B. At this time, two images are read, the first image being W_ODD and the second image being W_EVEN. W_ODD and W_EVEN are images corresponding to the sum of signals from the FPN of the X-ray imaging device 104 and X-rays. Therefore, by subtracting F_ODD from W_ODD and F_EVEN from W_EVEN, the FPN-removed images WF_ODD and WF_EVEN of the X-ray imaging apparatus 104 are obtained. This is called offset correction.

WF_ODD is an image corresponding to X-rays 302 (or high-energy X-rays 402) in the stable period. WF_EVEN is the sum of rising phase X-ray 301, stable phase X-ray 302, and falling phase X-ray 303 (or the sum of low energy X-rays 401 and 403 and high energy X-ray 402). is an image corresponding to Therefore, by subtracting WF_ODD from WF_EVEN, an image corresponding to the sum of the X-rays 301 in the rising period and the X-rays 303 in the falling period is obtained. The energies of the X-rays 301 in the rising period and the X-rays 303 in the falling period are lower than the energy of the X-rays 302 in the stable period. Therefore, by subtracting WF_ODD from WF_EVEN, a low energy image W_Low without a subject is obtained. Also, from WF_ODD, a high-energy image W_High with no subject is obtained. This is called color correction.

Next, an X-ray is emitted to the X-ray imaging device 104 in a state where an object is present to perform imaging, and an image is acquired by the driving shown in FIG. 3A or 3B. At this time, two images are read, the first image being X_ODD and the second image being X_EVEN. By performing the same subtraction as when there is no subject, a low energy image X_Low when there is a subject and a high energy image X_High when there is a subject are obtained.

Here, let d be the thickness of the subject, μ be the linear attenuation coefficient of the subject, I ₀ be the output of the pixel 20 when there is no subject, and I be the output of the pixel 20 when there is the subject. holds.

By transforming the formula (1), the following formula (2) is obtained. The right side of Equation (2) indicates the attenuation rate of the object. The attenuation rate of the subject is a real number between 0 and 1.

Therefore, by dividing the low-energy image X_Low with a subject by the low-energy image W_Low with no subject, an image L with an attenuation rate at low energy is obtained. Similarly, by dividing the high-energy image X_High with a subject by the high-energy image W_High with no subject, the image H of the attenuation rate at high energy is obtained. This is called gain correction.

Description of Signal Processing FIG. 4B shows a block diagram of signal processing in energy subtraction processing. In the signal processing, separated images are obtained from the attenuation rate image L at low energy and the attenuation rate image H at high energy obtained by the correction processing shown in FIG. 4A. Here, as an example of separated images, a case of obtaining a bone thickness image B (also referred to as a bone image B) and a soft tissue thickness image S (also referred to as a soft tissue image S) will be described.

Let E be the energy of X-ray photons, N(E) be the number of photons at energy E, B be the thickness of bone, and S be the thickness of soft tissue. Further, when the linear attenuation coefficient of bone at energy E is μ _B (E), the linear attenuation coefficient of soft tissue at energy E is μ _S (E), and the attenuation rate is I/I ₀ , the following equation (3) is It holds.

The number of photons N(E) at energy E is the X-ray spectrum. The X-ray spectrum is obtained by simulation or actual measurement. Also, the linear attenuation coefficient μ _B (E) of bone at energy E and the linear attenuation coefficient μ _S (E) of soft tissue at energy E can be obtained from databases such as NIST. That is, it is possible to calculate the attenuation rate I/I ₀ for any bone thickness B, soft tissue thickness S, X-ray spectrum N(E).

Here, assuming that the spectrum of low-energy X-rays is N _L (E) and the spectrum of high-energy X-rays is N _H (E), the following equation (4) holds.

By solving the nonlinear simultaneous equations of Equation (4), the thickness B of the bone and the thickness S of the soft tissue can be obtained. A case of using the Newton-Raphson method as a representative method for solving nonlinear simultaneous equations will be described. First, let m be the number of iterations of the Newton-Raphson method, Bm be the thickness of the bone after the ^mth iteration, and Sm be the thickness of the soft tissue after the ^mth iteration. The attenuation rate is H ^m , and the low-energy attenuation rate L ^m after the m-th iteration is represented by the following equation (5).

Also, the change rate of the attenuation rate when the thickness changes minutely is expressed by the following formula (6).

At this time, the bone thickness B ^m+1 and the soft tissue thickness S ^m+1 after the m+1th iteration are represented by the following equation (7) using the high-energy attenuation rate H ^m and the low-energy attenuation rate L ^m . .

The inverse matrix of a 2×2 matrix is represented by the following formula (8) from Cramer's formula, where det is the determinant.

Therefore, by substituting formula (8) into formula (7), the following formula (9) is obtained.

By repeating such calculations, the difference between the high-energy attenuation rate Hm after the ^m -th iteration and the actually measured high-energy attenuation rate H approaches zero limitlessly. The same is true for the attenuation rate L of low energy. As a result, the bone thickness Bm after the ^mth iteration converges to the bone thickness B, and the soft tissue thickness Sm converges to the soft tissue thickness S after the ^mth iteration. As described above, the nonlinear simultaneous equations shown in Equation (4) can be solved. Therefore, by calculating Equation (4) for all pixels, the bone thickness image B and the soft tissue thickness image B are obtained from the attenuation rate image L at low energy and the attenuation rate image H at high energy. S can be obtained.

In this embodiment, as an example of the separated image, an example of calculating the image of the thickness B of the bone and the thickness S of the soft tissue is shown, but the embodiment is not limited to such a form. For example, the thickness W of water and the thickness I of the contrast agent may be calculated as separate images. That is, it may be decomposed into thicknesses of any two types of materials. Further, an image of the effective atomic number Z and an image of the surface density D may be obtained as separate images from the image L of the attenuation rate at low energy and the image H of the attenuation rate at high energy obtained by the correction process of FIG. . The effective atomic number Z is the equivalent atomic number of a mixture, and the areal density D is the product of the object density [g/cm ³ ] and the object thickness [cm]. Of course, other operations may be performed on the low energy image and the high energy image to generate separated images. In other words, the signal processing of the present invention can be said to be processing for generating an energy subtraction image by computing a low energy image and a high energy image (energy subtraction processing). In this specification, the terms energy subtraction image and separation image are synonymous.

Also, in this embodiment, an example of solving the nonlinear simultaneous equations using the Newton-Raphson method is shown, but the method is not limited to such a form. For example, an iterative solution method such as the least squares method or the bisection method may be used. Also, in the present embodiment, the nonlinear simultaneous equations are solved by the iterative solution method, but the present invention is not limited to such a form. Bone thickness B and soft tissue thickness S for various combinations of high energy attenuation rate H and low energy attenuation rate L are obtained in advance to generate a table, and by referring to this table, bone thickness B and A configuration for obtaining the thickness S of the soft tissue at high speed may be used.

Description of Image Processing FIG. 4C shows a block diagram of image processing related to energy subtraction processing. In the image processing, an image for display is generated using the separated images obtained by the signal processing described above. For example, a display image is generated by post-processing the bone image B obtained by the signal processing shown in FIG. 4B. The generated display image is displayed on the display 145, for example. Such post-processing may include logarithmic transformation, dynamic range compression, and the like. The content of processing may be switched by inputting the type and strength of post-processing as parameters.

The energy subtraction processing according to this embodiment has been described above. FIG. 6 schematically shows an example of an accumulated image 601 and a bone image 602. As shown in FIG. The stored image 601 is an image before energy subtraction processing, that is, an image captured by an existing radiation imaging system without energy resolution or an image corresponding thereto. For example, image 306, image 406 in FIGS. 3A and 3B, high energy image H, and low energy image L described above correspond to accumulated images. Bone image 602 is a separated image obtained by the energy subtraction process described above. The normal human body consists only of soft tissue and bone. However, when Interventional Radiology (hereinafter referred to as IVR) is performed using the radiation imaging system shown in FIG. 1, a contrast agent is injected into the blood vessel. Treatments such as inserting a catheter or guide wire into a blood vessel and placing a stent or coil are also performed. In IVR, treatment is performed while confirming the positions and shapes of contrast agents and medical devices. Therefore, it is desired to improve the visibility by isolating only the contrast agent and medical device or by removing the background such as soft tissue and bone.

As shown in FIG. 6, in the accumulated image 601, the soft tissue is visible, whereas in the bone image 602 obtained by the energy subtraction processing, the contrast of the soft tissue can be removed. The main component of contrast media is iodine, and the main component of medical devices is metal such as stainless steel. Since both have atomic numbers greater than calcium, which is the main component of bone, the bone image 602 displays the bone, the contrast medium, and the medical device. As a result of studies conducted by the inventors of the present application, even if the high-energy image H and the low-energy image L are separated into a water image W and a contrast agent image I, bones, a contrast agent, and a medical device are included in the contrast agent image I. was confirmed to be displayed. The same is true for combinations of other two substances. The same effect can be obtained by changing the tube voltage and filter for low-energy X-rays and high-energy X-rays. In either case, it was confirmed that the bone image 602 displayed the bone, the contrast agent, and the medical device.

When the contrast of the soft tissue reduces the visibility, such as in the lung field when performing IVR of the chest, by displaying the bone image 602 in the radiation imaging system according to the first embodiment, contrast enhancement It may improve the visibility of drugs and medical devices. However, the bone image 602 has a larger frame-to-frame variation than the accumulated image 601, and there is a problem that the screen flickers when a moving image is displayed.

The pixel value of the accumulated image 601 varies for each shooting (frame). Possible causes include X-ray source-related factors such as variations in exposure dose and radiation quality between frames, and sensor-related factors such as sample-and-hold timing changes between frames. Variations in pixel values in an accumulated image are amplified by performing energy subtraction processing. FIG. 7 shows inter-frame variation with respect to the average image of frames before and after the energy subtraction process when radiographic moving images of porcine organs are captured. A graph 701 shows inter-frame variation in a moving image (bone image) after energy subtraction processing. A graph 702 represents inter-frame variation in the moving image (high energy image H) before the energy subtraction process. However, the image before the energy subtraction process adjusts the pixel value to [cm] by the following deformation.

Assuming that the X-rays are monochromatic and the material of the object is the same (the attenuation rate is the same), each pixel I/ _I0 of the image before energy subtraction can be expressed by equation (2) . However, μ is the attenuation rate at the average energy E of X-rays. Here, equation (2) can be transformed into equation (10) below. By transforming each pixel I/ _I0 of the image before the energy subtraction process as shown in Equation (10), it can be made to have the dimension of thickness [cm]. The bone attenuation rate is used as the attenuation rate μ.

As can be seen from FIG. 7, fluctuations in signal values between frames (graph 702), which are inconspicuous in the video before energy subtraction processing, become large in the video after energy subtraction processing (graph 701). Due to such a change in signal value between frames, an image blinks when a moving image is viewed. In this way, blinking may occur in the moving image after the energy subtraction processing, even if blinking is not visible in the moving image before the energy subtraction processing. Processing for reducing blinking in moving images after energy subtraction processing will be described below.

FIG. 8 shows a block diagram of signal processing including processing for reducing fluctuations (blinking) in signal values between frames (hereinafter, flickering reduction processing). The block diagram of FIG. 8 shows in more detail the signal processing blocks described above with reference to FIG. 4B. In the signal processing, separated images (bone image B′ and soft tissue A moving image of the tissue image S') is output. In the moving image of the separated image, flickering between frames is reduced by flickering reduction processing, which will be described later. Note that FIG. 8 shows the minimum required configuration for the sake of explanation. Also, block R1, block MD1, and block R2 can be realized by executing a program stored in ROM 142 by CPU 141, for example. Also, any one or all of the blocks may be implemented by dedicated hardware, or may be implemented by cooperation between the CPU and dedicated hardware.

First, the block R1 performs flicker reduction processing on the high energy image H and the low energy image L. Details of the flickering reduction process will be described later. Block MD1 then applies the signal processing described with reference to FIG. to generate Block R2 then performs a blink reduction process on bone image B and soft tissue image S to generate bone image B' and soft tissue image S'. As described above, from block R2, moving images B′ and S′, which are energy subtraction processed moving images subjected to flickering reduction processing, are obtained.

FIG. 9 is a flowchart showing the flickering reduction process performed by block R1 shown in FIG. In S901, block R1 calculates representative signal values from high energy image H and low energy image L, respectively. Preferably, the representative signal value is unaffected by subject motion. A representative signal value may be obtained from the entire image, or may be obtained from a set ROI. When setting the ROI, the ROI may be a predetermined fixed region, may be set manually by the user when calculating the representative signal value, or may be set automatically. As a method of automatically setting the ROI, for example, a method of analyzing a moving image to specify an area where there is no signal value variation due to the object, and setting the specified area as the ROI can be used. Also, as the representative signal value, a statistic such as the median value, the average value, or the N% value of the cumulative histogram can be used. In the following description, the median value is used as the representative value signal.

In S902, the block R1 obtains a reference representative signal value (hereinafter referred to as reference value) from the moving image data to be subjected to flickering reduction processing. The reference value can be, for example, the median value of the first frame image of the moving image data. More specifically, the median value of the image of the first frame of the moving image data composed of the series of high energy images H is used as the reference value for the high energy image, and the median value of the image of the first frame of the moving image data composed of the series of low energy images L The median value of the image is determined as the reference value for the low energy image.

Next, in S903, the block R1 calculates the correction amount for each of the high energy image H and the low energy image L. The correction amount α of the high-energy image H and the correction of the high-energy image H using the correction amount α will be described below, but the low-energy image can also be corrected by similar processing. The correction amount α is obtained by calculating how many times the reference value obtained from the moving image data of the high energy image in S902 is the median value obtained for the high energy image H in S901. At S904, block R1 multiplies the high energy image H by the correction amount α obtained at S903 to obtain a high energy image H′. In this way, the frame-to-frame variation of the signal in the moving image composed of the high-energy images H is reduced. As an example of processing for reducing inter-frame variation, a calculation for obtaining a high-energy image H′ subjected to blink reduction processing from a high-energy image H is shown in Equation (11). However, in Equation (11), H[N] indicates the Nth frame of the moving image made up of the high-energy image H, and MED(H[N]) indicates the median value of the Nth frame of the moving image. Equation (11) is an example of blinking reduction processing using multiplication of a coefficient (correction amount α).

By repeating the above processing until the end of shooting (S905), moving image data of the high-energy image H' with reduced inter-frame variation of the signal (reduced blinking) is obtained. Also, by similar processing, a moving image of the low-energy image L′ with reduced inter-frame variation of the signal (reduced blinking) is obtained.

In the above example, the reference value was obtained from the first frame image, but it is not limited to this. For example, the reference value may be calculated using all or part of the frames captured at the time the frame to be corrected was obtained (that is, the frames prior to the frame to be corrected). For example, the reference value may be calculated from the latest average image of a predetermined number of frames at the time when the frame to be corrected is acquired.

Also, in the above description, the correction amount is calculated independently for the high energy image and the low energy image, but the present invention is not limited to this. You can do it. For example, the high-energy image H and the low energy image L may be corrected.

For example, for ΔH and ΔL in equation (12), equation (13) shows a case where the correction amount is calculated so that ΔH=ΔL=(ΔH+ΔL)/2.

Also, in the above example, the correction amount is obtained by division, and the correction is performed by multiplication, but the present invention is not limited to this. For example, the correction amount may be derived and corrected by addition/subtraction. Equation (14) shows the case of obtaining a high energy image H′ subjected to blink reduction processing from the high energy image H by addition/subtraction. Equation (14) is an example of blink reduction processing using addition and subtraction of a coefficient (correction amount α).

The flickering reduction processing for correcting the images before the energy subtraction processing (the high energy image H and the low energy image L) has been described above. The flickering reduction processing performed by the block R1 on images (high energy image H and low energy image L) corresponding to radiography with a plurality of energies before separation of correction targets has been described above. A block R2 performs flickering reduction processing with the separated image as the target of correction. A method similar to the above-described flickering reduction processing (for example, the method represented by Equation (11) or Equation (14)) can be used for the flickering reduction processing.

FIG. 10 shows the frame-to-frame variation of the image average value in the moving image obtained by the energy subtraction processing with and without the above-described flickering reduction processing. A graph 701, like FIG. 7, shows the variation of the image average value in the bone image when the flicker reduction processing is not performed. A graph 1001 shows the variation of the image average value in the bone image when the flicker reduction processing is performed. As is clear from FIG. 10, the inter-frame variation of the average value is suppressed by the flickering reduction processing, and the flickering in the moving image is reduced.

As described above, according to the first embodiment, flickering in a moving image obtained by energy subtraction processing is reduced, and an easy-to-observe image is obtained.

It should be noted that the flickering reduction process used for block R1 and block R2 may be either a method of multiplying coefficients or a method of adding or subtracting coefficients. However, as a result of investigation by the inventor of the present application, blinking can be suppressed most effectively when multiplication correction is performed in blinking reduction processing for the high energy image H and the low energy image L. FIG. In addition, flickering can be suppressed most effectively when the bone image B and the soft tissue image S are corrected by addition and subtraction in the flickering reduction process. Therefore, it is preferable to apply correction by multiplication of coefficients to the blink reduction processing of block R1 and correction by addition and subtraction of coefficients to the blink reduction processing of block R2. Further, in the above embodiment, the flicker reduction process is performed on both the image before the energy subtraction process and the image after the energy subtraction process, but the present invention is not limited to this. At least one of block R1 and block R2 may perform the blink reduction process. However, when the inventors of the present application examined the block R1 and the block R2, it was found that blinking could be reduced more satisfactorily by using both the block R1 and the block R2.

(Second embodiment)
In the second embodiment, in addition to the bone image flickering reduction processing described in the first embodiment, a configuration for performing noise reduction processing for reducing noise in the bone image will be described.

The noise reduction processing of this embodiment effectively reduces noise using stored images that are compatible with images acquired by general radiography. First, an accumulated image used for noise reduction processing according to this embodiment will be described. As described in the first embodiment, for one X-ray irradiation, high-resolution images are obtained from a plurality of radiographic images acquired by sample-holding at a plurality of timings including timings during X-ray irradiation and after the end of X-ray irradiation. An energy image and a low energy image are generated. Here, for example, an X-ray image (image 306 in FIG. 3A) acquired at the timing after the end of X-ray irradiation can be used as the accumulated image.

FIG. 11 is a block diagram showing an example of a configuration for implementing correction processing for acquiring an accumulated image according to the second embodiment. The accumulated image A is generated, for example, by dividing the image XF_EVEN by the image WF_EVEN. Image XF_EVEN and image WF_EVEN are as described in FIG. 4A. That is, the image XF_EVEN is an image corresponding to the sum of the X-rays 301 in the rising period, the X-rays 302 in the stable period, and the X-rays 303 in the falling period when there is an object. The image WF_EVEN is an image corresponding to the sum of the X-rays 301 in the rising period, the X-rays 302 in the stable period, and the X-rays 303 in the falling period when there is no subject.

Note that the stored image A may be generated by multiplying an image H with an attenuation rate in high energy (high energy image H) and an image L with an attenuation rate in low energy (low energy image L) and adding them together. good. For example, accumulated image A may be generated using equation (15). Note that in calculating the accumulated image A, one coefficient may be set to 0 and the other coefficient may be set to 1. In this case, the high energy image H or the low energy image L itself is used as the accumulated image A. That is, an image captured at substantially the same timing as the image to be subjected to energy subtraction processing and to which energy subtraction processing has not been applied can be used as accumulated image A for noise reduction processing.

FIG. 12 is a block diagram showing an example of a configuration for performing signal processing in energy subtraction processing according to the second embodiment. The block diagram of FIG. 12 shows a more detailed configuration of the signal processing described with reference to FIG. 4B. Note that FIG. 12 shows the minimum required configuration for the sake of explanation. Blocks R1-R3, blocks MD1-MD2, block ADD, and blocks F1-F3 can be implemented by the CPU 141 executing a program stored in the ROM 142, for example. However, any one or all of the blocks shown in FIG. 12 may be implemented by dedicated hardware, or may be implemented by cooperation between the CPU and dedicated hardware.

Block F2 and block F3 perform filtering for the purpose of noise reduction on the low energy image L and the high energy image H, respectively. Generate image H'. For filtering, for example, spatial filters such as Gaussian filters and median filters, structure-preserving spatial filters such as epsilon filters and Laplacian filters, and time filters such as recursive filters can be used. Quantum noise of the X-rays is reduced by a noise reduction process that applies a filter before the block MD1 that performs two-material separation. Block R1 performs a blink reduction process on the noise reduced low energy image L' and the noise reduced high energy image H' to produce a high energy image H'' and a low energy image L''. Note that the blink reduction processing by block R1 is the same as the blink reduction processing by the first embodiment (block R1 in FIG. 8). A moving image composed of the high-energy image H'' and the low-energy image L'' is a moving image in which inter-frame fluctuations of signals are suppressed, both of which are obtained from the block R1.

Next, a block MD1 generates a bone image B' and a soft tissue image S' from the high energy H'' and low energy image L'' that have undergone noise reduction processing and are corrected for inter-frame variation. The operation of material separation by block MD1 is the same as in the first embodiment (block MD1 in FIG. 8). Next, a block R2 performs flicker reduction processing on the bone image B' and the soft tissue image S' generated in the block MD1, and outputs a bone image B'' and a soft tissue image S''. The processing by block R2 is the same as in the first embodiment (block RD2 in FIG. 8).

A block ADD generates an image of the sum of the bone image B'' and the soft tissue image S'', and outputs it as a thickness image T'. At this stage, the thickness image T', which is the sum of the bone image B'' and the soft tissue image S'', has undergone noise reduction and blink reduction. A block F1 applies filter processing for the purpose of noise reduction to the thickness image T' to generate a thickness image T'' after the filter processing. The thickness image T'' is a thickness image to which flicker reduction and noise reduction have been applied in a double manner. Spatial filters such as Gaussian filters and median filters, structure-preserving spatial filters such as epsilon filters and Laplacian filters, and temporal filters such as recursive filters can be used for the filtering process of block F1.

A block H1 generates an accumulated image A'' using, for example, equation (15) from the high energy image H'' and the low energy image L'' subjected to the blink reduction process, which are output from the block R1. A block MD2 performs material separation processing from the thickness image T'' output from the block F1 and the accumulated image A'' output from the block H1, and performs flickering reduction processing and noise reduction processing to obtain a bone image. B''' is generated. The material separation processing by block MD2 will be described later. A block R3 performs further flickering reduction processing on the bone image B''' obtained by the block MD2 to generate a flickering-reduced bone image B''''. According to the noise reduction processing using the block F1 and the block MD2 as described above, noise accompanying the energy subtraction processing (substance separation) is reduced.

Next, the substance separation processing by block MD2 will be described. Under the assumption (constraint condition) that the accumulated image consists only of bone and soft tissue, let _A be the pixel value of the accumulated image, NA (E) be the spectrum in the accumulated image, S be the thickness of the soft tissue, and S be the thickness of the bone. Assuming that the thickness of is B, the following equation (16) holds.

Here, if the sum of the thickness of the bone and the thickness of the soft tissue is T, the following formula (17) holds by transforming the formula (16) from T=B+S.

By substituting the pixel value A and the thickness T of the accumulated image at a certain pixel into Equation (17) and solving the nonlinear equation, it is possible to obtain the thickness B of the bone at a certain pixel. That is, if the thickness image T′ subjected to noise reduction processing and flickering reduction processing is used as the thickness T, and the accumulated image A″ output from the block H1 is used as the pixel value A of the accumulated image, Equation (17) is solved. , the bone thickness B representing the bone image B''' is obtained. Since the thickness image has higher continuity than the accumulated image, it does not contain high frequency components. Therefore, even if noise is removed by filtering the thickness image T' using the block F1, the signal component is less likely to be lost. The noise-reduced bone image B''' can be obtained by the block MD2 using the noise-reduced thickness image T'' and the accumulated image A'', which originally has little noise. Similarly, by transforming equation (16) with B=TS, block MD2 obtains a noise-reduced soft tissue image S''' from the thickness image T'' and the accumulated image A''. It is also possible to

Furthermore, in the above description, the image T'' synthesized with the block MD2 is used, but it is not limited to this. For example, by substituting the pixel values of the soft-tissue image obtained by noise-reducing the soft-tissue image S″, which is the output of the block R2, into the equation (16), the block MD2 performs noise reduction and blink reduction. A bone image B''' obtained by performing the above may be obtained. Alternatively, by substituting the pixel values of the bone image obtained by noise-reducing the bone image B'' output from the block R2 in the block F1 into the equation (16), the block MD2 is subjected to noise reduction and blink reduction. A soft tissue image S''' may be obtained.

Through the above processing, according to the second embodiment, it is possible to generate an energy subtraction video in which inter-frame fluctuations (blinking) of signals are reduced and noise is reduced.

Note that in the second embodiment, not all of the noise reduction processes described above need to be executed. For example, the noise reduction processing by the filters of block F2 and block F3 may be omitted. Alternatively, for example, the noise reduction processing by the filter in block F1 may be omitted.

It should be noted that the flickering reduction processing used for block R1, block R2, and block R3 may be either a method of multiplying coefficients or a method of adding or subtracting coefficients. However, as a result of investigation by the inventor of the present application, blinking can be suppressed most effectively when multiplication correction is performed in blinking reduction processing for the high energy image H and the low energy image L. FIG. In addition, flickering could be suppressed most effectively when addition and subtraction were performed in flickering reduction processing for the bone image B' and the soft tissue image S'. Therefore, it is preferable to apply correction by multiplication to the blink reduction processing of block R1, and correction by addition and subtraction to the blink reduction processing of blocks R2 and R3. Further, in the above embodiment, the flickering reduction process is performed in blocks R1 to R3, but the present invention is not limited to this. At least one of block R1, block R2, and block R3 may be subjected to the blink reduction process. However, according to the study of the present inventors, it is preferable from the viewpoint of blinking reduction to perform the blinking reduction processing by the block R1 and the block R2.

When performing double noise reduction as shown in FIG. 11, filter blocks F2 and F3 applied before block MD1 for two-substance separation, and filter block F1 applied after block MD1, It is necessary to optimize the type and strength at the same time. This is because the result of optimizing the two filters independently is not always optimal. For example, if a temporal direction filter or a spatial direction filter is applied twice, the increase rate of the quantum noise of X-rays and the noise accompanying the separation of two substances will not be independent, and the two noise reduction effects may not be integrated. be. Thus, for example, blocks F2 and F3, which are filters applied before block MD1 for two-substance separation, apply filters in the temporal direction, and filters applied after MD1, block F1, apply filters in the spatial direction. A configuration can be used. Of course, it is also possible to apply a spatial filter in blocks F2 and F3 and a temporal filter in block F1.

It should be noted that it is also possible to adopt a configuration in which either the filter in the spatial direction or the filter in the temporal direction is doubled, and in that case, it is preferable to make the size of the kernel and the size of the filter coefficients different for both. For example, when filtering in the spatial direction twice, it is preferable to configure the filter kernel of block F1 to be larger than the kernels of blocks F2 and F3. This is because the thickness image T has higher spatial continuity than the accumulated image A, the high energy image H, and the low energy image L, for example. Moreover, when applying filters in the time direction twice, it is preferable to configure the filter coefficient of the block F1 to be larger than the filter coefficients of the filters of the blocks F2 and F3. This is because, for example, the thickness image T changes less over time than the accumulated image A, the high energy image H, and the low energy image L.

In addition, block F1 may be configured to apply both temporal and spatial filters, and blocks F2 and F3 may be configured to simultaneously apply both temporal and spatial filters. In any configuration, if a spatial or temporal filter is doubly applied before and after the two-matter separation block MD1, a configuration with larger filter coefficients or kernels in the filter block F1 after MD1. can be preferably used.

In addition, in the first and second embodiments, the X-ray imaging device 104 is an indirect radiation sensor using a phosphor, but is not limited to such a form. For example, a direct radiation sensor using a direct conversion material such as CdTe may be used. In addition, in the first and second embodiments, the passive tube voltage change of the X-ray generator 101 is used (FIG. 3A), or the tube voltage is actively switched (FIG. 3B). is not limited to any form. The energy of the radiation irradiated to the X-ray imaging device 104 may be changed by switching the filter of the X-ray generation device 101 over time.

Also, in the first and second embodiments, the case of two X-ray (radiation) energies has been described. However, the present invention is not limited to such a form, and the above-described embodiment can be applied even when the X-ray (radiation) has three or more energies. By performing energy subtraction processing using n images corresponding to n energies (n is a natural number of 2 or more), n material separation images are obtained. By applying the above-described blink reduction processing to n images (moving images) before energy subtraction processing and/or to n separated images (moving images) after energy subtraction processing, separated images with reduced blinking are obtained. .

Furthermore, in the first and second embodiments, energy subtraction was performed by changing the energy of the radiation irradiated to the X-ray imaging device 104, but the present invention is not limited to such a form. For example, by stacking two two-dimensional detectors 106 (sensors), a method of changing the spectrum of radiation detected by the front sensor and the rear sensor may be used. Alternatively, a plurality of images with different energies may be obtained by using a photon counting sensor that counts the number of radiation quanta for each energy.

Also, in the first and second embodiments, energy subtraction processing was performed using the control computer 103 of the radiation imaging system, but the present invention is not limited to such a form. For example, the control computer 103 may be incorporated into the X-ray imaging device 104 . Alternatively, the image acquired by the control computer 103 may be transferred to another computer to perform energy subtraction processing. For example, a configuration in which an acquired image is transferred to another personal computer (image viewer) via a medical PACS and displayed after energy subtraction processing is preferably used. That is, in each of the above embodiments, it is sufficient to provide radiation images with different energies to the energy subtraction process, and the method for acquiring radiation images with different energies is not limited to the above embodiments. In the above embodiment, the control computer 103 directly acquires an image from the X-ray imaging apparatus 104 and performs energy subtraction processing, but the present invention is not limited to this. A moving image captured by the X-ray imaging apparatus 104 may be stored in an external storage device, and the control computer 103 may read out the moving image from the storage device and perform energy subtraction processing.

As described above, according to each of the above-described embodiments, it is possible to provide an image processing apparatus or radiation imaging system capable of creating an energy subtraction moving image in which flicker between frames is suppressed.

(Other examples)
The present invention supplies a program that implements one or more functions of the above-described embodiments to a system or device via a network or a storage medium, and one or more processors in the computer of the system or device reads and executes the program. It can also be realized by processing to It can also be implemented by a circuit (for example, ASIC) that implements one or more functions.

The invention is not limited to the above embodiments, and various changes and modifications are possible without departing from the spirit and scope of the invention. Accordingly, the claims are appended to make public the scope of the invention.

This application claims priority based on Japanese Patent Application No. 2021-030625 submitted on February 26, 2021, and the entire contents of the description are incorporated herein.

101: X-ray generator, 102: X-ray controller, 103: Control computer, 104: X-ray generator

Claims (18)

1. generating means for generating moving images of separated images by performing energy subtraction processing using a plurality of moving images corresponding to a plurality of different radiation energies;
   correction means for correcting the plurality of moving images or the moving image of the separated images so as to reduce variation in signal values between frames in the moving images of the plurality of moving images or the moving image of the separated images;
   An image processing device comprising:
2. The image processing apparatus according to claim 1, wherein the correcting means corrects the frame of the moving image to be corrected based on a statistic calculated based on frames previous to the frame.
3. The image processing apparatus according to claim 2, wherein the correcting means calculates the statistic from pixel values of all or part of each frame.
4. The image processing apparatus according to claim 3, wherein the statistic is any one of the median value, the average value, and the N% value of the cumulative histogram.
5. 5. The image processing apparatus according to claim 3, further comprising setting means for specifying an area in which there is no change in signal value due to an object by analyzing the moving image to be corrected, and setting the area as the partial area. .
6. The correcting means is
   Obtaining a statistic calculated based on a specific frame in the moving image to be corrected as a reference value,
   6. The image processing apparatus according to any one of claims 2 to 5, wherein each frame is corrected based on the statistic calculated from each frame of the moving image to be corrected and the reference value.
7. The image processing apparatus according to claim 6, wherein the correction means acquires, as the reference value, a statistic calculated from the first frame image of the moving image to be corrected.
8. The correcting means acquires, as the reference value, a statistic calculated from an average image obtained from a predetermined number of latest frames among frames past the frame for each frame of the moving image to be corrected. The image processing apparatus according to claim 6.
9. 7. The correcting means multiplies the pixel value of the frame by a coefficient based on a ratio between the statistic calculated from the frame and the reference value in correcting the plurality of moving images or correcting the separated image moving image. 9. The image processing apparatus according to any one of items 1 to 8.
10. wherein the correcting means adds or subtracts a coefficient based on a difference between a statistic calculated from a frame and the reference value to or from a pixel value of the frame in correcting the plurality of moving images or correcting the separated image moving image. Item 10. The image processing apparatus according to any one of Items 6 to 9.
11. The correcting means is
    For each of the plurality of moving images, correction is performed by multiplying the pixel value of the frame by a coefficient based on the ratio of the statistic calculated from the frame and the reference value,
    9. The moving image of the separated image is corrected by adding or subtracting a coefficient based on the difference between the statistic calculated from the frame and the reference value to or from the pixel value of the frame. The image processing device according to .
12. 9. Any one of claims 6 to 8, wherein the correcting means corrects each frame for each of the plurality of moving images so that the difference between the statistic and the reference value in each frame is the same. 10. The image processing device according to claim 1.
13. Further comprising a first noise reduction means for reducing noise in the plurality of moving images,
    13. The image processing apparatus according to any one of claims 1 to 12, wherein the correction means corrects the plurality of moving images whose noise has been reduced by the first noise reduction means.
14. Further comprising second noise reduction means for reducing frame noise in the moving image of the separated image based on an accumulated image obtained by weighted addition of the frames of the plurality of moving images,
    The correcting means corrects at least one of the moving image of the separated image before being processed by the second noise reducing means and the moving image of the separated image after being processed by the second noise reducing means. 14. The image processing apparatus according to any one of claims 1 to 13, to be corrected.
15. 15. The image processing apparatus according to claim 14, wherein the second noise reduction means obtains the accumulated image by performing weighted addition of the plurality of moving image frames corrected by the correction means.
16. imaging means for capturing a plurality of moving images corresponding to a plurality of different radiation energies;
    generating means for generating a moving image of separated images by performing energy subtraction processing using the plurality of moving images;
    correction means for correcting the plurality of moving images or the moving image of the separated images so as to reduce variation in signal values between frames in the moving images of the plurality of moving images or the moving image of the separated images;
    A radiation imaging system comprising:
17. a generation step of generating a moving image of separated images by performing energy subtraction processing using a plurality of moving images corresponding to a plurality of different radiation energies;
    a correction step of correcting the moving images of the plurality of moving images or the moving images of the separated images so as to reduce variations in signal values between frames in the moving images of the plurality of moving images or the moving images of the separated images;
    An image processing method including
18. A program that causes a computer to function as each means of the image processing apparatus according to any one of claims 1 to 15.

Images