Radiographic apparatus, radiographic system, radiographic method, and program

Abstract

To provide a radiographic apparatus and the like capable of obtaining an optimum still / dynamic image, even when an unexpected change occurs in a patient's movement and the like during radiographing of an object site and the like.SOLUTION: A console 3 comprises: an image acquisition part that acquires a radiation image of a radiographic site of a subject; a positional information acquisition part that acquires positional information on the subject on the basis of an optical image of the subject photographed by at least an optical camera; and a correction part that corrects the radiation image on the basis of the positional information on the subject during the radiographing of the radiation image.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to a radiation imaging apparatus, a radiation imaging system, a radiation imaging method, and a program.

Background Art

[0002] In recent years, in radiation imaging apparatuses, in addition to taking still images and diagnosing them, it has become possible to take dynamic images for confirming the movement of a target site and diagnose them. When taking still images and dynamic images, the patient is required to take positions and movements according to the imaging purpose.

[0003] A technique for confirming the patient's positioning and the movement of the target site using a camera different from the radiation imaging apparatus has been proposed. Patent Document 1 describes an imaging support apparatus that acquires the movement of the abdomen of a subject with a camera before the start of imaging and evaluates the stability of the cycle of the dynamics of the target site based on the acquired biological information.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the conventional technology, in order to confirm the patient's positioning and movement before the start of radiation imaging, there is a problem that if there are unexpected changes in the patient's movement or the like during radiation imaging of the target site or the like, desired still images and dynamic images cannot be obtained.

[0006] Therefore, an object of the present invention is to provide a radiation imaging apparatus, a radiation imaging system, a radiation imaging method, and a program that can obtain optimal still images and dynamic images even when there are unexpected changes in the patient's movement or the like during imaging of the target site or the like.

Means for Solving the Problem

[0007] The radiographic apparatus according to the present invention includes an image acquisition unit that acquires a radiographic image of a target part of a subject, a position information acquisition unit that acquires at least the position information of the subject, and a correction unit that corrects the radiographic image based on the position information of the subject during the shooting of the radiographic image. It is provided with.

[0008] The radiographic imaging system according to the present invention includes an image acquisition unit that acquires a radiographic image of a target part of a subject, a position information acquisition unit that acquires at least the position information of the subject, and a correction unit that corrects the radiographic image based on the position information of the subject during the shooting of the radiographic image. It is provided with.

[0009] The radiographic imaging method according to the present invention includes an image acquisition step of acquiring a radiographic image of a target part of a subject, a position information acquisition step of acquiring at least the position information of the subject, and a correction step of correcting the radiographic image based on the position information of the subject during the shooting of the radiographic image. It has.

[0010] The program according to the present invention causes a computer to function as an image acquisition unit that acquires a radiographic image of a subject, a position information acquisition unit that acquires at least the position information of the subject, and a correction unit that corrects the radiographic image based on the position information of the subject during the shooting of the radiographic image. function as such.

Effect of the Invention

[0011] According to the present invention, since the radiographic image is corrected based on the position information of the subject during the imaging of the radiographic image, even if there is an unexpected change in the movement of the patient or the like during the imaging of the target part or the like, an optimal still image and dynamic image can be obtained.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Embodiment for Carrying Out the Invention

[0013] [Configuration Example of Radiation Imaging System 100] FIG. 1 is a diagram showing an example of the schematic configuration of a radiation imaging system 100 according to the present embodiment. The radiation imaging system 100 includes a radiation generator 1, a radiation image capturing device 2, a console 3, an image management device 4, and a HIS / RIS 5. Hereinafter, the radiation image capturing device 2 may be referred to as the capturing device 2, and the radiation generator 1 may be referred to as the generator 1. The generator 1, the capturing device 2, the console 3, the image management device 4, and the HIS / RIS 5 are communicably connected to each other via a network N. Examples of the network N include a LAN, a WAN, or the Internet. LAN is an abbreviation for Local Area Network. WAN is an abbreviation for Wide Area Network. The communication method of the network N may be wired communication or wireless communication.

[0014] The generator 1 includes a generator 11, an exposure switch 12, and a radiation source 13. The generator 11 applies a voltage corresponding to preset imaging conditions to a radiation source 13 including, for example, an X-ray tube based on the operation of the exposure switch 12. The generator 11 may have an operation unit that receives an input such as irradiation conditions.

[0015] When a voltage is applied from the generator 11, the radiation source 13 generates radiation R having a dose corresponding to the applied voltage. The radiation R is, for example, X-ray.

[0016] The generating device 1 is provided with an optical camera 14. The optical camera 14 is arranged, for example, at a position adjacent to the radiation source 13. For example, the optical camera 14 may be mounted in the same way as the radiation source 13. The optical camera 14 photographs the subject S and its surrounding area simultaneously or substantially simultaneously with the photographing of the radiation image of the subject S. The optical image includes a still image or a moving image continuously photographed. The optical camera 14 transmits the photographed optical image to the console 3. Note that the optical camera 14 may transmit the photographed optical image to the console 3 after performing predetermined processing on the photographed optical image with another device.

[0017] The generating device 1 generates radiation R in a mode corresponding to the type of radiation image, for example, a still image or a dynamic image. Specifically, in the case of a still image, the generating device 1 irradiates the radiation R only once when the exposure switch 12 is pressed once. In the case of a dynamic image, the generating device 1 performs dynamic imaging in which, in response to one shooting operation, radiation such as X-rays is pulsed and repeatedly irradiated to the subject S at predetermined time intervals, thereby acquiring a series of images of the subject S. Repeatedly irradiating the radiation in pulses at predetermined time intervals is called pulse irradiation. Also, dynamic imaging includes a case where, in response to one shooting operation, a series of images of the subject S are acquired by irradiating the subject S continuously at a low dose rate without interruption. Continuously irradiating the radiation without interruption is called continuous irradiation. A series of images obtained by dynamic imaging is called a dynamic image. Also, each of all the images constituting the dynamic image is called a frame image. Dynamic imaging includes video shooting, but does not include shooting a still image while displaying a video. Also, the dynamic image includes a video, but does not include an image obtained by shooting a still image while displaying a video.

[0018] The imaging device 2 generates digital image data in which the imaging part of the subject S is shown. For example, a portable FPD is used for the imaging device 2. FPD is an abbreviation for Flat Panel Detector. Note that the imaging device 2 may be integrally configured with the generating device 1.

[0019] Although not shown in the drawings, the imaging device 2 includes, for example, an imaging element, a sensor substrate, a scanning unit, a reading unit, a control unit, and a communication unit. The imaging element generates charges corresponding to the dose by receiving radiation R. The sensor substrate has switch elements arranged two-dimensionally (in a matrix) and accumulates and discharges charges. The scanning unit switches on / off each switch element. The reading unit reads the amount of charges released from each pixel as a signal value. The control unit generates image data of a radiation image from a plurality of signal values read by the reading unit. The image data includes still image data or moving image data. The communication unit transmits the generated image data and various signals, etc. to other devices such as the console 3, and receives various information and various signals from other devices.

[0020] The console 3 sets imaging conditions for the generator 1, the imaging device 2, etc., and controls the reading operation of the radiation image captured by the imaging device 2. The console 3 is also called an imaging control device and is configured by, for example, a personal computer or the like.

[0021] The imaging conditions include, for example, patient conditions regarding the subject S, irradiation conditions regarding the irradiation of the radiation R, and image reading conditions regarding the image reading of the imaging device 2, etc. The patient conditions are, for example, the imaging site, the imaging direction, the physique, etc. The irradiation conditions are, for example, the tube voltage (kV), the tube current (mA), the irradiation time (ms), the current-time product (mAs value), etc. The image reading conditions are, for example, the pixel size, the image size, and the frame rate, etc. The console 3 may automatically set the imaging conditions based on order information obtained from the HIS / RIS 5 or the like. Also, the console 3 may be configured to manually set the imaging conditions by an operation of an operation unit 31 by a user such as a doctor or a radiological technologist, for example.

[0022] The image management device 4 stores and manages the radiation image data generated by the imaging device 2, the optical image data generated by the optical camera 14, the data subjected to dynamic analysis processing, etc. The image management device 4 outputs the radiation image data, etc. to an information terminal, a viewer, etc. based on an output instruction by the user. The image management device 4 may be referred to as a PACS. PACS is an abbreviation for Picture Archiving and Communication System.

[0023] The HIS / RIS 5 receives, for example, order information regarding the radiation imaging of a patient from a doctor or the like, and transmits the received order information to the console 3. The order information includes various types of information such as, for example, the patient ID, the imaging site, the imaging direction, the build, etc.

[0024] [Configuration example of console 3] Next, the configuration of the console 3 according to the present embodiment will be described. FIG. 2 is a diagram showing an example of a block diagram of the console 3 according to the present embodiment.

[0025] The console 3 includes a control unit 30, an operation unit 31, a display unit 32, a storage unit 33, and a communication unit 34. The control unit 30, the operation unit 31, the display unit 32, the storage unit 33, and the communication unit 34 are connected to each other via wiring such as a bus 35, for example.

[0026] The control unit 30 has, for example, a processor such as a CPU that performs arithmetic operations and control, and a memory. The CPU is an abbreviation for Central Processing Unit. The control unit 30 realizes a process of correcting the positional deviation (movement deviation) of a radiation image using an optical image during the imaging of the radiation image, for example, by executing a program P described later stored in a memory such as a RAM, the storage unit 33, etc. Note that the control unit 30 may include an electronic circuit such as an ASIC or an FPGA. The ASIC is an abbreviation for Application Specific Integrated Circuit. The FPGA is an abbreviation for Field Programmable Gate Array.

[0027] In this embodiment, the control unit 30 functions as an image acquisition unit, a position information acquisition unit, and a correction unit. The processor or the like of the control unit 30 realizes the functions of the image acquisition unit, the position information acquisition unit, and the correction unit by executing the program P stored in the storage unit 33 or the like. The image acquisition unit acquires a radiation image of a still image or a moving image of the subject S. The position information acquisition unit acquires at least the position information of the subject S. The position information of the subject S may be the skeleton points of the optical image of the subject S. The correction unit corrects the radiation image based on the position information of the subject S during the shooting of the radiation image by a process based on machine learning or an affine transformation. The correction unit corrects the frame of the moving image according to the timing when the amount of movement of the position information of the subject S becomes equal to or greater than a preset threshold value.

[0028] The operation unit 31 includes, for example, a mouse, a keyboard, a switch, a button, etc. The operation unit 31 may be, for example, a touch panel integrally combined with a display, or an interface that accepts voice input. The operation unit 31 receives an instruction according to various input operations from the user, converts the received instruction into an operation signal, and outputs it to the control unit 30. Specifically, the operation unit 31 receives a selection instruction or the like as to which of the radiation images before and after the correction of the positional deviation is to be used as the inspection image.

[0029] The display unit 32 is, for example, a display such as a liquid crystal display or an organic EL display. EL is an abbreviation for Electro Luminescence. The display unit 32 displays an image on which a predetermined analysis process has been performed, a GUI or the like for receiving various input operations from the user. GUI is an abbreviation for Graphical User Interface. Specifically, the display unit 32 displays the radiation image of the imaging part of the subject S photographed by the imaging device 2, and also displays an optical image or the like including the subject S photographed by the optical camera 14.

[0030] The storage unit 33 includes, for example, any storage module such as an HDD, SSD, ROM, and RAM. HDD is an abbreviation for Hard Disk Drive. SSD is an abbreviation for Solid State Drive. ROM is an abbreviation for Read Only Memory. The storage unit 33 stores, for example, system programs, application programs, and various data. Specifically, the storage unit 33 stores a program P for executing processes such as correcting the positional deviation of a radiation image using an optical image when the movement of the subject S exceeds the assumption. When a data group such as a radiation image taken at a timing when the subject S of the optical image has a movement exceeding the assumption is input to the storage unit 33, a learned model M that outputs a radiation image with the positional deviation corrected is stored. In this embodiment, the learned model M is stored in the storage unit 33 of the console 3, but it may be stored in an external device such as a cloud server, for example.

[0031] The communication unit 34 includes, for example, a communication module including a NIC, a receiver, and a transmitter. NIC is an abbreviation for Network Interface Card. The communication unit 34 communicates various information and image data with the generator 1, the imaging device 2, the image management device 4, the HIS·RIS 5, etc. via the network N.

[0032] [Operation example of console 3] Next, the radiation imaging method according to this embodiment will be described. FIG. 3 is a flowchart showing an example of the operation of the console 3 during dynamic imaging according to this embodiment. In this embodiment, a dynamic image of the imaging site of the subject S is acquired by dynamic imaging using the generator 1 and the imaging device 2, and the case of obtaining a radiation image of the chest as the imaging site will be described.

[0033] The control unit 30 acquires an optical image including the subject S photographed by the optical camera 14, and also acquires a dynamic image of the photographed part of the subject S photographed by the photographing device 2 (step S1). Step S1 corresponds to an image acquisition step. The optical camera 14 can perform a wider-angle photographing than the photographing device 2. Therefore, the optical image photographed by the optical camera 14 includes images of the photographing device 2 around the subject S, the background wall, etc. in addition to the photographed part of the subject S. The optical camera 14 starts photographing a moving image in accordance with the start timing of dynamic photographing of the photographing device 2. The photographing of the radiation image is started by turning on the exposure switch 12.

[0034] The control unit 30 acquires the position information of the subject S from the acquired optical image (step S2). Step S2 corresponds to a position information acquisition step. For the position information of the subject S, for example, the skeleton points of the subject S in the optical image are used. The skeleton points of the subject S can be acquired by a skeleton detection process. The skeleton detection process may use, for example, a known application (software) capable of detecting predetermined skeleton points from the subject. Also, a learned model obtained by performing machine learning for detecting predetermined skeleton points from the optical image may be used. In this case, the control unit 30 inputs the optical image photographed by the optical camera 14 into the learned model to acquire the skeleton points of the subject S from the optical image. Further, the control unit 30 may acquire position information other than the skeleton points of the subject S.

[0035] The skeleton points of the subject S are composed of, for example, 18 locations. Specifically, the 18 skeleton points are: (1) nose, (2) neck, (3) right shoulder, (4) right elbow, (5) right wrist, (6) left shoulder, (7) left elbow, (8) left wrist, (9) right hip, (10) right knee. The other skeleton points are: (11) right ankle, (12) left hip, (13) left knee, (14) left ankle, (15) right eye, (16) left eye, (17) right ear, (18) left ear. Each skeleton point includes the coordinate position (x, y) on the plane of the optical image. For each skeleton point, a reliability indicating certainty may be estimated. The reliability can be quantified from 0.0 to 1.0. The closer to 1.0, the higher the reliability.

[0036] In this embodiment, when photographing a photographing site specified by order information or the like, skeletal points preset according to combinations such as the type of photographing and the photographing site are acquired. That is, when photographing a predetermined photographing site, the control unit 30 does not acquire all 18 skeletal points, but only acquires the skeletal points necessary for each photographing site. Thereby, the speed of the skeleton detection process can be increased. Note that information regarding the photographing site may be acquired from sources other than order information. For example, an application or the like that can identify the photographing site from an optical image or the like may be used.

[0037] FIG. 4 is a diagram showing the correspondence between the type of photographing, the photographing site, and the skeletal points of the subject S according to this embodiment. For example, when the type of photographing is respiratory photographing and the photographing site is the chest, (3) right shoulder, (6) left shoulder, (9) right waist, and (12) left waist are acquired as skeletal points. Even when the type of photographing is respiratory photographing and the photographing site is the abdomen, (3) right shoulder, (6) left shoulder, (9) right waist, and (12) left waist are acquired as skeletal points, as in the case of the chest. When the type of photographing is plastic surgery photographing and the photographing site is the neck, (1) nose, (2) neck, (3) right shoulder, (6) left shoulder, (15) right eye, (16) left eye, (17) right ear, and (18) left ear are acquired as skeletal points. When the type of photographing is plastic surgery photographing and the photographing site is the right elbow, (3) right shoulder, (4) right elbow, and (5) right wrist are acquired as skeletal points. When the type of photographing is plastic surgery photographing and the photographing site is the left knee, (12) left waist, (13) left knee, and (14) left ankle are acquired as skeletal points.

[0038] FIG. 5 is a diagram showing an example of the skeletal points of the subject S in the optical image I1 detected by the skeletal detection process when the type of imaging is respiratory imaging and the imaging site is the chest. The patient is positioned in a standing posture with the chest facing the imaging device 2 side. The control unit 30 acquires four skeletal points of the right shoulder N3, left shoulder N6, right waist N9, and left waist N12 from the subject S in the optical image I1 by executing the skeletal detection process. Note that the control unit 30 may also acquire position information other than the skeletal points corresponding to the imaging type and imaging site shown in FIG. 5. Specifically, the control unit 30 may acquire position information of the imaging device 2, peripheral device I3, background wall I4, etc. shown in the optical image I1. Hereinafter, the imaging device 2, peripheral device I3, background wall I4, etc. may be collectively referred to as peripheral information In. Thereby, the three-dimensional movements such as the front-back direction and twisting direction of the subject S can be confirmed more accurately.

[0039] The control unit 30 determines whether or not the amount of movement of the skeletal points of the subject S in the acquired optical image is equal to or greater than a preset threshold value (step S3). Step S3 corresponds to the correction step. The control unit 30 calculates the amount of movement of the skeletal points of the acquired optical image by comparing the skeletal points of the reference optical image with the skeletal points of the subject S in the acquired optical image. The reference optical image is, for example, an optical image when the subject S is positioned directly in front of the imaging device 2 and is in an ideal positioning. The reference optical image may be acquired from the optical image during imaging or from a past optical image or the like. The control unit 30 compares the calculated amount of movement with the preset threshold value. The threshold value is set, for example, based on the amount of movement of the subject S when the subject S has moved significantly more than expected with respect to the imaging device 2 and a dynamic image necessary for diagnosis cannot be obtained. Further, the control unit 30 may add the peripheral information In to the skeletal point information of the subject S to detect the movement of the subject S.

[0040] When the control unit 30 determines that the amount of movement of the skeletal points of the subject S in the optical image is equal to or greater than a preset threshold value, it proceeds to step S4. The control unit 30 corrects the frame of the radiation image according to the timing when the amount of movement of the skeletal points of the subject S in the optical image becomes equal to or greater than the threshold value (step S4). This is because when there is more movement of the subject S in the optical image than expected, there is also a displacement of the position of the imaging region in the frames of the dynamic image captured at the same timing. Note that the same timing includes timings with a slight deviation. Also, when the frame rates of the optical image and the radiation image are made to match, the timing when the amount of movement of the skeletal points of the subject S in the optical image becomes equal to or greater than the threshold value may be managed by the frame rate.

[0041] The control unit 30 extracts the frame of the dynamic image captured at the timing when the amount of movement of the skeletal points of the subject S in the optical image becomes equal to or greater than a preset threshold value. The control unit 30 corrects the displacement of the extracted frame of the dynamic image. Hereinafter, the extracted frame of the dynamic image may be referred to as the target radiation image. FIG. 6 is a diagram showing a correction method for correcting the displacement of the frame of the extracted dynamic image according to the present embodiment using the learned model M. The control unit 30 inputs a data group including the optical image, the radiation image, etc. to the learned model M. Examples of the data group include a reference optical image I1 when the subject S is in an ideal positioning, and the reference skeletal points of the reference optical image I1. Examples of other data groups include a target optical image I2 when the amount of movement of the skeletal points of the subject S becomes equal to or greater than the threshold value, the target skeletal points of the target optical image I2, and the target radiation image G1 corresponding to the target optical image I2. The control unit 30 obtains, as an output result output from the learned model M in response to the input of the data group, a radiation image G2 in which the displacement of the imaging region is corrected. In the radiation image G2, the amount of movement of the skeletal points of the subject S that has become equal to or greater than the threshold value is offset, and the two-dimensional displacement of the imaging region and the three-dimensional displacements in the front-rear direction and the torsion direction of the imaging region are corrected.

[0042] Here, a method for constructing the learned model M will be described. The learned model M may be constructed by providing a learning function to a console 3 or the like, or may be created by another external device such as a computer having a learning function. Hereinafter, the device for constructing the learned model M is referred to as a learning device. FIG. 7 is a diagram showing a construction method when constructing the learned model M according to the present embodiment. The learning device generates the learned model M by performing machine learning using a plurality of data sets of data with a positioning deviation of the subject S and data with no positioning deviation of the subject S, that is, ideal positioning data. For example, an optical image including the positioning deviation of the subject S, the skeleton points of the subject S in the optical image, and data of a radiation image corresponding to the optical image are used in the data set. As other data sets, for example, an optical image with no positioning deviation of the subject S, the skeleton points of the subject S in the optical image, and data of a radiation image corresponding to the optical image are used. By using not only the radiation image including the imaging site of the subject S but also the optical image including the subject S and the surrounding information In as learning data, it is possible to learn not only the two-dimensional movement of the subject S but also the movement of the subject S in the depth direction. As a result, from the learned model M, it is possible to output a radiation image in which not only the two-dimensional positioning deviation but also the three-dimensional positioning deviation at the imaging site is corrected. As a method for constructing the learned model M, a known method such as a neural network can be used.

[0043] Returning to FIG. 3, the control unit 30 displays on the screen of the display unit 32 the radiation image obtained by dynamic imaging and the radiation images before and after correction when the positional deviation is corrected (step S5). FIG. 8 is a diagram showing a display example 1 of the screen of the display unit 32 on which the radiation images before and after correction according to the present embodiment are displayed. In the following description, the display of the radiation image obtained by dynamic imaging is omitted for convenience. When the control unit 30 corrects the positional deviation of the radiation image of a predetermined frame, the radiation image G1 before correction and the radiation image G2 after correction are displayed side by side on the screen of the display unit 32. By checking the screen of the display unit 32, the user can visually confirm at which frame of the dynamic image and to what extent the positional deviation has been corrected.

[0044] FIG. 9 is a diagram showing a display example 2 of the screen of the display unit 32 on which the radiation images before and after correction according to the present embodiment are displayed. When the control unit 30 corrects the positional deviation of the radiation image of a predetermined frame, the radiation image G1 before correction and the radiation image G2 after correction are displayed overlapping on the screen of the display unit 32. It is preferable that the radiation image G1 before correction and the radiation image G2 after correction have transparency with respect to each other. Thereby, even when the radiation image G1 before correction and the radiation image G2 after correction are displayed overlapping, both images can be visually recognized. Also, in FIG. 9, the radiation image G1 before correction and the radiation image G2 after correction are displayed overlapping, but either one of the radiation image G1 before correction and the radiation image G2 after correction may be switched and displayed. For example, each time the image is clicked, the radiation image G1 before correction and the radiation image G2 after correction may be switched and displayed.

[0045] FIG. 10 is a diagram showing a third display example of the screen of the display unit 32 on which the radiation images before and after correction according to the present embodiment are displayed. When the control unit 30 corrects the positional deviation of the radiation image of a predetermined frame, the control unit 30 extracts the lung field region R1 of the radiation image G1 before correction and the lung field region R2 of the radiation image G2 after correction, respectively. The control unit 30 overlays and displays the extracted lung field region R1 before correction and the lung field region R2 after correction on the screen of the display unit 32. For example, the control unit 30 displays the region where the lung field region R1 before correction and the lung field region R2 after correction overlap in red, the region where only the lung field region R1 before correction exists in blue, and the region where only the lung field region R2 after correction exists in green. In FIG. 10, each region of red, blue, and green is shown with a difference in shading.

[0046] FIG. 11 is a diagram showing a fourth display example of the screen of the display unit 32 on which the radiation images before and after correction according to the present embodiment are displayed. The control unit 30 sequentially arranges and displays each frame of the dynamic image acquired from the imaging device 2 from the first frame at the lower part of the screen of the display unit 32. At this time, the control unit 30 highlights the frame with a large correction amount. As the highlighting, the frame may be surrounded by a predetermined color, for example, a red frame and displayed. Thereby, the user can intuitively visually recognize the magnitude of the correction amount for each frame in the dynamic image. Here, the magnitude of the correction amount can be determined as follows. The control unit 30 determines that the correction amount is large when the total value of the difference values between the pixels of the radiation image G1 before correction and the radiation image G2 after correction exceeds a preset threshold value. On the other hand, the control unit 30 determines that the correction amount is small when the total value of the difference values between the pixels of the radiation image G1 before correction and the radiation image G2 after correction is less than a preset threshold value. The difference value can be calculated by, for example, the mean squared error. When the frame highlighted by the user is selected, the control unit 30 enlarges and displays the radiation image G1 before correction and the radiation image G2 after correction of the selected frame side by side at the substantially central part of the screen. A button 22a for selecting which of the radiation image G1 before correction and the radiation image G2 after correction to adopt is displayed on the screen of the display unit 32.

[0047] FIG. 12 is a diagram showing a display example 5 of the screen of the display unit 32 on which radiation images before and after correction according to the present embodiment are displayed. The control unit 30 sequentially arranges and displays each frame of the moving image acquired from the imaging device 2 at the lower part of the screen of the display unit 32 starting from the first frame. When a predetermined frame is selected by the user, the control unit 30 enlarges and displays the pre-correction radiation image G1 and the post-correction radiation image G2 of the selected frame side by side in the substantially central part of the screen. The control unit 30 displays the correction content of the post-correction radiation image G2 with respect to the pre-correction radiation image G1 in the correction value display unit 32b on the right side of the screen. Examples of the display content of the correction value display unit 32b include the amount of movement in the left-right direction, the amount of movement in the up-down direction, the amount of rotation in the left-right direction, and the amount of rotation in the up-down direction. Each correction content of the correction value display unit 32b can be calculated by the method shown below. The control unit 30 maps the skeleton points of the subject S acquired by the optical image to each of the pre-correction radiation image G1 and the post-correction radiation image G2. The control unit 30 calculates the difference between the coordinates of the skeleton points of the pre-correction radiation image G1 after the mapping process and the coordinates of the skeleton points of the post-correction radiation image G2, thereby calculating the amount of movement in the left-right direction, the amount of movement in the up-down direction, and the like.

[0048] Returning to FIG. 3, in step S3, when the control unit 30 determines that the movement of the skeleton points of the subject S in the acquired optical image is not greater than a preset amount of movement, the process proceeds to step S6. That is, when there is no movement of the subject S beyond the assumption and an optimal moving image necessary for diagnosis is obtained. The control unit 30 displays the moving image dynamically captured by the imaging device 2 on the screen of the display unit 32 (step S6).

[0049] In the above-described embodiment, the radiation image with a positional shift in the imaging region due to unexpected movement of the subject S is corrected by a process based on machine learning, but the radiation image may be corrected by other correction methods. For example, the control unit 30 can also correct the radiation image with a positional shift in the imaging region by a process based on affine transformation such as translation and rotation. Since the flow of the correction process is common except for step S4 of the flowchart shown in FIG. 3, detailed description of the common parts is omitted.

[0050] When the control unit 30 determines that the amount of movement of the skeletal points of the subject S in the optical image is equal to or greater than the threshold value, it extracts the frame of the dynamic image captured at the same timing as this optical image. For example, the control unit 30 performs an affine transformation so as to align the extracted frame with a specific frame among the frames during dynamic shooting that has no misalignment or the misalignment is within the allowable range. In the affine transformation, the coordinates of each frame are transformed using a matrix. Examples of the specific frame include the first frame, a frame adjacent to the extracted frame, and the like. Further, when calculating the amount of movement of the skeletal points of the subject S in the optical image with respect to the reference, the amount of movement may be used as a correction value, and the misalignment of the frames of the dynamic image may be corrected based on the correction value.

[0051] As another correction method, when the patient is under follow-up observation in the diagnosis result and there is a past dynamic image of the same imaging site of the same patient, correction processing may be performed using the past dynamic image. In this case, the control unit 30 selects a frame of the past dynamic image that has no misalignment or the misalignment is within the allowable range. The control unit 30 performs an affine transformation so as to align the extracted frame of the dynamic image with the selected frame of the past dynamic image.

[0052] Also, in the above-described embodiment, the case where the radiation image is a dynamic image has been described, but the correction processing according to this embodiment can also be applied when the radiation image is a still image. For example, when the patient is under follow-up observation in the diagnosis result and there is a past still image of the same imaging site of the same patient, correction processing may be performed using the past still image. In this case, the control unit 30 selects a still image that has no misalignment or the misalignment is within the allowable range in the still image captured in the past. When the misalignment of the skeletal points of the subject S in the optical image is equal to or greater than the threshold value, the control unit 30 performs an affine transformation so as to align the radiation image of the still image captured at the same timing as this optical image with the selected past still image. As another correction method, as described above, a learned model capable of outputting a still image in which the misalignment at the imaging site is corrected may be used.

[0053] As described above, in the present embodiment, during the shooting of a radiation image such as a moving image, it is determined whether or not there is an unexpected change in the movement of the subject S based on, for example, the skeleton points of the subject S in the optical image. When there is an unexpected change in the movement of the subject S in the optical image or the like, correction is performed on the radiation image taken at the same timing as this optical image so as to cancel out the change in the movement amount of the subject S. Thereby, according to the present embodiment, even when there is an unexpected change in the movement of the subject S during the shooting of the radiation image, a still image or a moving image suitable for diagnosis can be obtained. Further, according to the present embodiment, even when the radiation image is a moving image, since the correction process and the display process are executed in real time, appropriate processing can also be executed for the subsequent dynamic analysis process.

[0054] Further, in the present embodiment, a learned model M is constructed using an optical image I1 or the like including the subject S and the surrounding information In, and the radiation image including the positional deviation of the imaging site is corrected using this learned model M. Therefore, in addition to the two-dimensional movement of the subject S in the vertical and horizontal directions, a radiation image corrected for the rotational and torsional movements of the subject S in the front-rear direction and the left-right direction can be obtained. Thereby, only the planar positional deviation of the imaging site could be corrected with only the radiation image, but by using the optical image I1 or the like, the three-dimensional positional deviation of the imaging site can be corrected.

[0055] Further, according to the present embodiment, the radiation image G1 and the radiation image G2 before and after correction, the correction content, etc. are displayed on the screen of the display unit 32. Thereby, even when a user such as a radiological technologist executes a correction process based on machine learning, the user can confirm which frames of the moving image have been corrected. Further, according to the present embodiment, when there is a sense of incongruity in the corrected frame, it is also possible to return to the frame before correction.

[0056] The preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings. However, the technical scope of the present disclosure is not limited to such examples. Also, various modified examples and improvements naturally fall within the technical scope of the present disclosure within the scope of the technical idea described in the claims of those skilled in the art.

[0057] In the above embodiment, an example has been described in which when the subject S of the optical image has an unexpected movement, the radiation image corresponding to the optical image is corrected in real time, and the result is displayed on the display unit 32. However, the present disclosure is not limited to this. For example, after the acquisition of all the radiation images is completed, the correction process of the radiation image using the above-described optical image may be performed. Also, in the correction process based on the learned model M, specific optical image and radiation image data are input, but the present disclosure is not limited to this. For example, all the optical images and radiation images obtained by imaging may be input to the learned model M. In this case, the learned model M performs, for example, a correction process on the radiation image with misalignment and outputs all the input radiation images.

Explanation of Reference Numerals

[0058] 3 Console (Radiographic Apparatus) 30 Control Unit 100 Radiographic System G1 Radiation Image, Target Radiation Image G2 Radiation Image I1 Optical Image I2 Target Optical Image M Learned Model S Subject

Claims

1. An image acquisition unit that acquires a radiation image of a target part of a subject; A position information acquisition unit that acquires at least the position information of the subject; A correction unit that corrects the radiation image based on the position information of the subject during the shooting of the radiation image; A radiation imaging apparatus comprising the above.

2. The radiation image is a dynamic image, The correction unit corrects a frame of the dynamic image according to a timing when a movement amount of the position information of the subject becomes equal to or greater than a preset threshold value. The radiation imaging apparatus according to Claim 1.

3. The correction unit corrects the radiation image by processing based on machine learning or affine transformation. The radiation imaging apparatus according to Claim 1.

4. The position information acquisition unit acquires the position information based on an optical image of the subject, The correction unit inputs at least the optical image in which the movement amount of the position information of the subject becomes equal to or greater than a preset threshold value and the radiation image corresponding to the optical image into a learned model, and acquires from the learned model the radiation image in which the positional deviation of the target part is corrected. The radiation imaging apparatus according to Claim 3.

5. The correction unit corrects the frame of the dynamic image so as to match a specific frame different from the frame of the dynamic image. The radiation imaging apparatus according to Claim 2.

6. The correction unit corrects the frame of the dynamic image so as to match a specific frame of a past dynamic image at the same imaging part of the same patient. The radiation imaging apparatus according to Claim 2.

7. The position information acquisition unit acquires the position information based on an optical image of the subject. The radiation imaging apparatus according to Claim 1.

8. The correction unit acquires the position information by detecting skeleton points of the subject in the optical image. The radiation imaging apparatus according to Claim 7.

9. An image acquisition unit that acquires a radiation image of a target part of a subject; A position information acquisition unit that acquires at least the position information of the subject; A correction unit that corrects the radiation image based on the position information of the subject during the shooting of the radiation image; A radiation imaging system comprising the above.

10. An image acquisition step of acquiring a radiation image of a target part of a subject; A position information acquisition step of acquiring at least the position information of the subject; A correction step of correcting the radiation image based on the position information of the subject during the shooting of the radiation image; A radiation imaging method comprising the same. **Claim 11** A computer, An image acquisition unit that acquires a radiation image of a subject, A position information acquisition unit that acquires at least the position information of the subject, A correction unit that corrects the radiation image based on the position information of the subject during the shooting of the radiation image, A program for causing the computer to function as such.

Images