X-ray imaging apparatus and image processing method

Abstract

The system detects the body movement of the subject while fluoroscopic images are being taken. [Solution] An image of the subject is captured with a camera. The skeleton of the human figure contained in the image is detected, and based on the detected skeleton, the skeleton of the subject mounted on the top plate of the X-ray imaging device is extracted. The amount of movement (optical flow) of the corresponding pixels is calculated between the captured image and an image acquired by the camera at a predetermined time before the timing of acquisition of the captured image. Multiple regions of interest are set in the image of the subject based on the position of the subject's skeleton. The determination unit determines whether body movement has occurred in the subject based on the amount of movement of the subject's pixels within the regions of interest. The region of interest selection unit includes a region of interest selection unit, which either selects one or more regions of interest from the multiple regions of interest set by the region of interest selection unit to be used by the determination unit for body movement determination, or specifies the position of the region of interest to be used by the determination unit for body movement determination to the region of interest selection unit.

Description

[[Technical Field]]

[0001] The present invention relates to an X-ray imaging apparatus that acquires images by irradiating a subject with X-rays. [[Background Art]]

[0002] IVR (Interventional Radiology) is known, in which an endoscope or a catheter is inserted into a patient to perform various procedures on the patient while acquiring fluoroscopic images by irradiating the patient with X-rays from an X-ray imaging apparatus. A physician performs the procedure on the patient while checking a monitor displaying fluoroscopic images and a monitor displaying endoscopic images.

[0003] When checking the patient's condition during a procedure performed while viewing fluoroscopic images, the physician moves their line of sight to the patient's face, feet, or hands for visual confirmation, or moves their line of sight to a monitor that displays images from a camera mounted on the ceiling or upper wall of the room to check the patient via the camera images.

[0004] On the other hand, Patent Document 1 discloses a technique for detecting body movement of a subject by performing image processing on fluoroscopic images.

[0005] Patent Document 2 discloses a technique for identifying a worker performing a specific procedure from among a plurality of workers by capturing images of the plurality of working workers with a camera, processing the images, and performing skeleton detection. [[Prior Art Documents]] [[Patent Documents]]

[0006] [[Patent Document 1]] Japanese Unexamined Patent Application Publication No. 2021-186199 [[Patent Document 2]] Japanese Unexamined Patent Application Publication No. 2021-196783 [[Summary of the Invention]] [[Problem to be Solved by the Invention]]

[0007] Interventional radiology (IVR), which involves inserting an endoscope or catheter into a patient while acquiring fluoroscopic images and performing various procedures on the patient, is performed with the patient under sedation. However, patients may unconsciously try to pull out the endoscope or move their bodies. If body movement occurs, the physician must call a nurse or other staff member inside or outside the examination room and give instructions such as holding the patient down while performing the procedure. Therefore, physicians need to be aware of any movement of the patient even while intently watching the endoscope monitor during the procedure.

[0008] However, during surgery, doctors need to shift their gaze from the endoscope monitor to the patient in order to visually check the patient's condition, which is burdensome for the doctor.

[0009] On the other hand, when a doctor checks the patient's condition by looking at a monitor showing camera footage from inside the room, it is difficult to instantly grasp the patient's condition because the camera footage includes not only the patient but also doctors, technicians, and nurses.

[0010] The objective of this invention is to detect body movements of a subject while fluoroscopic images are being taken. [Means for solving the problem]

[0011] To achieve the above objective, the present invention provides an X-ray imaging apparatus comprising a top plate on which a subject is mounted, an X-ray irradiation device that irradiates the subject with X-rays, an X-ray detector that detects X-rays irradiated from the X-ray irradiation device and transmitted through the subject, and a motion detection unit that receives an image from a camera that images the subject, processes the image, and detects the occurrence of motion in the subject. The motion detection unit includes a skeleton detection unit, an optical flow calculation unit, a region of interest selection unit, and a determination unit. The skeleton detection unit detects the skeleton of a human figure included in the image and extracts the skeleton of the subject mounted on the top plate based on the detected skeleton. The optical flow calculation unit calculates the amount of movement of corresponding pixels between the image and an image acquired by the camera at a predetermined time before the timing of acquisition of the said image. The region of interest selection unit sets multiple regions of interest in the image of the subject based on the position of the subject's skeleton extracted by the skeleton detection unit. The determination unit determines whether motion is occurring in the subject based on the amount of movement of the subject's pixels within the region of interest. The region of interest selection unit includes a region of interest selection unit, which either selects one or more regions of interest from among a plurality of regions of interest set by the region of interest selection unit to be used by the determination unit for body movement determination, or the determination unit specifies the location of the region of interest to be used for body movement determination to the region of interest selection unit. [Effects of the Invention]

[0012] According to the present invention, it is possible to detect the body movements of a subject while fluoroscopic images are being taken. By informing the physician of the detection results, the physician can easily grasp when the patient's body movements occur, even during a procedure. [Brief explanation of the drawing]

[0013] [Figure 1] (a) Front view of the X-ray imaging apparatus 1 according to Embodiment 1 of this embodiment, (b), (c) Diagrams showing examples of images displayed on the camera image display monitor. [Figure 2] Functional block diagram of the X-ray imaging apparatus of Embodiment 1. [Figure 3] A diagram illustrating an example of the arrangement of the top plate 80 of the X-ray imaging device 1, the subject 10, and the doctors and nurses in a typical examination room. [Figure 4] Fig. 1 is a diagram showing an image captured by a camera of the X-ray imaging apparatus 1 according to Embodiment 1 and a skeleton detected from the image. [Figure 5] Fig. 4 is a diagram showing a region of interest 52 set at the feature points of the skeleton in Fig. 4 according to Embodiment 1. [Figure 6] (a) is an image with pixel values representing optical flow (movement amount) calculated from an image captured by a camera of the X-ray imaging apparatus according to Embodiment 1, (b) is an image showing moving shadow pixels calculated from an image captured by the camera, and (c) is a diagram illustrating exclusion of a region of interest containing a large number of shadow pixels. [Figure 7] Fig. 7 is a flowchart showing the operation of the X-ray imaging apparatus according to Embodiment 1. [Figure 8] Fig. 11 is a diagram showing a state where a notification dialog 161 for notifying occurrence of body movement and a graph 162 are displayed on the display screen of the camera image display monitor 15 of the X-ray imaging apparatus according to Embodiment 1. [Figure 9] Fig. 12 is an enlarged view of the graph 162 in Fig. 11. [Figure 10] (a) and (b) are diagrams illustrating notification of occurrence of body movement based on the edge color of an image 21a when the image 21a is displayed on the display screen of the camera image display monitor 15 of the X-ray imaging apparatus according to Embodiment 1. [Figure 11] (a) is a diagram showing an image of a subject 10 partially covered with a protective curtain and the set region of interest 52 according to Embodiment 2-1, and (b) is a diagram showing an image from which one region of interest 52a has been excluded. [Figure 12] Fig. 15 is a functional block diagram of the X-ray imaging apparatus according to Embodiment 2-1. [Figure 13] Fig. 16 is a flowchart showing the operation of the X-ray imaging apparatus according to Embodiment 2-1. [Figure 14] (a) is a diagram showing that the protective curtain presence button of the X-ray imaging apparatus according to Embodiment 2-1 is off and the set region of interest, and (b) is a diagram showing that the protective curtain presence button of the X-ray imaging apparatus according to Embodiment 2-1 is on and the set region of interest. [Figure 15] Fig. 18 is a functional block diagram of the X-ray imaging apparatus according to Embodiment 2-2. [Figure 16] Fig. 19 is a flowchart showing the operation of the X-ray imaging apparatus according to Embodiment 2-2. [Figure 17] FIG. 1 is a functional block diagram of the X-ray imaging apparatus according to Embodiment 2-3. [Figure 18] (a) is an image showing a whole body selection button 204a of the X-ray imaging apparatus according to Embodiment 2-3 and a set region of interest, (b) is an image showing an upper body selection button 204b of the X-ray imaging apparatus according to Embodiment 2-3 and a set region of interest, and (c) is an image showing a head and shoulders only selection button 204c of the X-ray imaging apparatus according to Embodiment 2-3 and a set region of interest. [Figure 19] FIG. 1 is a flowchart showing the operation of the X-ray imaging apparatus according to Embodiment 2-3. [Figure 20] FIG. 1 is an image showing a second region of interest 153 of the X-ray imaging apparatus according to Embodiment 3-1, a subject 10, and a skeleton other than the subject 10. [Figure 21] FIG. 1 is a flowchart showing the operation of the X-ray imaging apparatus according to Embodiment 3-1. [Figure 22] (a) is an image showing a second region of interest 153 of the X-ray imaging apparatus according to Embodiment 3-1, a subject 10, and a skeleton other than the subject 10, and (b) is an image showing the subject 10 and skeleton lines other than the subject 10 of the X-ray imaging apparatus according to Embodiment 3-2. [Figure 23] FIG. 1 is a flowchart showing the operation of the X-ray imaging apparatus according to Embodiment 3-2. [Figure 24] FIG. 1 is an image showing a state where a skeleton other than the subject passes through the region of interest of the subject 10 of the X-ray imaging apparatus according to Embodiment 3-3. [Figure 25] FIG. 1 is a flowchart showing the operation of the X-ray imaging apparatus according to Embodiment 3-3. [Figure 26] (a) is a schematic diagram showing an image of the subject 10 of the X-ray imaging apparatus according to Embodiment 4 covered with a protective curtain, and (b) is a schematic diagram showing a state where a region of interest 52 is set in the image of the X-ray imaging apparatus according to Embodiment 4 covered with the protective curtain. [Figure 27] FIG. 1 is a functional block diagram of the X-ray imaging apparatus according to Embodiment 4. [Figure 28] FIG. 1 is a flowchart showing the operation of the X-ray imaging apparatus according to Embodiment 4. MODE FOR CARRYING OUT THE INVENTION

[0014] Hereinafter, an X-ray imaging apparatus according to an embodiment of the present invention will be described with reference to the drawings. The X-ray imaging device of this embodiment is equipped with a camera and has a function to determine whether or not the subject is moving based on the image captured by the camera. Specifically, the system first detects the skeleton of a human figure, and then, based on the detected skeleton, extracts the skeleton of the subject mounted on the top plate of the X-ray imaging device. Next, the optical flow is determined by calculating the amount of movement of corresponding pixels between the current image and an image acquired by the camera at a predetermined time before the current image was acquired. Furthermore, multiple regions of interest are set in the image of the subject based on the position of the subject's skeleton. One or more regions of interest are selected from the set regions of interest to be used for motion detection, or the location of one or more regions of interest to be used for motion detection is specified, and a region of interest is set at that location. Whether or not motion is occurring in the subject is determined using the amount of pixel movement (optical flow) of the selected region of interest or the region of interest set at the specified location. The motion detection function of the X-ray imaging apparatus of this embodiment will be further described below with reference to Embodiments 1 to 4.

[0015] <<Embodiment 1>> The X-ray imaging device 1 of Embodiment 1 will be described with reference to the drawings. Figure 1(a) is a front view of the X-ray imaging device 1, and Figures 1(b) and (c) are images displayed on the camera image display monitor. Figure 2 is a functional block diagram of the X-ray imaging device 1. Figure 3 is a diagram showing the arrangement of the X-ray imaging device, the subject, and the doctor and nurse in the examination room. Figures 4 to 6 are examples of images captured by the camera and image processing. Figure 7 is a flowchart showing the operation of the X-ray imaging device 1, and Figures 8 to 10 are diagrams showing examples of the display of the body motion detection results.

[0016] As shown in Figure 1(a), the X-ray imaging apparatus 1 comprises a top plate 80 on which a subject 10 is mounted, and an X-ray irradiation device 11 for irradiating the subject 10 with X-rays. The X-ray irradiation device 11 is supported on a stand 70 by an X-ray support unit 50. The stand 70 is equipped with an X-ray vertical movement mechanism 60 for moving the X-ray support unit 50 up and down, a longitudinal movement mechanism (not shown) for moving the X-ray support unit 50 in the longitudinal direction 50a of the top plate 80, a longitudinal movement mechanism (not shown) for moving the X-ray support unit 50 in the short direction 50b of the top plate 80, and an X-ray tilt mechanism 61 for tilting (50c) the central axis of the X-ray support unit 50 around the short direction of the top plate. The X-ray rotation mechanism (not shown) is also provided within the X-ray support unit 50 for rotating the X-ray irradiation device 11 around the longitudinal direction of the top plate. The X-ray support unit 50 is equipped with a pressing device 62 used to press on the abdomen of the subject 10 during imaging.

[0017] On the other hand, the top plate 80 is supported by the top plate support part 81 on the stand 70.

[0018] An X-ray detector 12 is located inside the top plate support section 81. The X-ray detector 12 is a planar detector in which X-ray detection elements are arranged in two dimensions. The X-ray detector 12 detects X-rays that are irradiated from the X-ray irradiation device 11 and have passed through the subject 10.

[0019] The X-ray support unit 50 or the X-ray irradiation device 11 is equipped with a camera 21 for imaging the subject 10. The camera 21 is positioned so as not to obstruct the X-rays emitted from the X-ray irradiation device 11. The camera 21 is set to capture the entire body of the subject 10 on the top plate 80, as shown in Figures 1(b) and (c). The camera 21 may also be fixed to the cover (housing) of the X-ray irradiation device 11.

[0020] The field of view of the camera 21 moves in accordance with the movement of the X-ray support unit 50 in the longitudinal direction 50a of the top plate 80, as shown in Figure 1(b), and moves in accordance with the movement of the X-ray support unit 50 in the short direction 50b of the top plate 80, as shown in Figure 1(c).

[0021] As shown in Figure 2, an X-ray image generation unit 13 is connected to the X-ray detector 12. The X-ray image generation unit 13 receives signals output by each X-ray detection element of the X-ray detector 12 when it detects X-rays, and generates an X-ray image (in this case, a fluoroscopic image) at a predetermined frame rate. A fluoroscopic image display monitor 14 is connected to the X-ray image generation unit 13. The fluoroscopic image is displayed on the fluoroscopic image display monitor 14.

[0022] Meanwhile, a motion detection unit 20 is connected to the camera 21. The motion detection unit 20 receives images from the camera 21 and processes them. Specifically, the motion detection unit 20 receives images from the camera 21 and detects the skeletons of all human figures contained in the received images. Based on the detected skeletons, the motion detection unit 20 calculates predetermined feature quantities and, based on the calculated feature quantities, extracts the image of the subject 10 mounted on the top plate 80 from among the multiple human figures. The motion detection unit 20 displays the extracted image of the subject 10 on the connected camera image display monitor 15.

[0023] More specifically, the motion detection unit 20 includes an image acquisition unit 22, an image rotation unit 23, a skeleton detection unit 24, a subject skeleton extraction unit 25, a region of interest selection unit 26, an optical flow calculation unit 27, a shadow detection unit 201, and a determination unit 28. The region of interest selection unit 26 also includes a region of interest selection unit 126.

[0024] The image acquisition unit 22 acquires the image 21a output by the camera 21 at predetermined time intervals. The predetermined time interval can be set to be approximately the same as the frame rate of the fluoroscopic images generated by the X-ray image generation unit 13. The image 21a output by the camera 21 is an image taken so that the body axis of the subject on the tabletop 80 is the horizontal axis of the image, as shown in Figures 1(b) and (c).

[0025] The image rotation unit 23 rotates image 21a by 90 degrees in a predetermined direction so that the position of the subject's head on the top plate 80 is at the top of image 21a, as shown in Figure 4. Since it is predetermined which end of the top plate 80 in the longitudinal direction the subject's head will be mounted on, the direction in which image 21a is rotated is predetermined by the predetermined position of the head on the top plate 80 and the orientation of the camera 21.

[0026] Furthermore, as shown in Figure 3, an endoscope is positioned near the tabletop 80 of the examination room, and medical personnel such as doctors and nurses are standing or sitting around the tabletop 80 in order to insert the endoscope into the subject 10 and perform the procedure. Therefore, as shown in Figures 1(b) and (c), the image 21a taken by the camera 21 includes not only the subject 10 but also the medical personnel.

[0027] The skeleton detection unit 24 extracts skeletal feature points (both ears, nose, both eyes, both shoulders, both elbows, both wrists, both hips, both knees, both ankles, etc.) of all human figures in image 21a, as shown in Figure 4. The skeleton detection unit 24 detects the skeleton of all human figures by detecting their two-dimensional coordinates. The skeleton detection unit 24 can be implemented using a known posture detection algorithm. For example, the posture estimation algorithm Open Pose®, included in Open VINO® provided by Intel Corp., can be used.

[0028] The subject skeleton extraction unit 25 uses the feature points of the skeletons of all the human figures detected by the skeleton detection unit 24 to calculate predetermined feature quantities, and extracts the skeleton of the subject 10 mounted on the top plate based on the calculated feature quantities.

[0029] For example, the subject skeleton extraction unit 25 identifies subjects from all detected human figures whose skeletons include shoulders and ears. For each subject candidate, the subject skeleton extraction unit 25 determines the midpoint 41 of the line connecting the subject candidate's shoulders, as shown in Figure 4, and calculates the distance between the determined midpoint 41 of the shoulders and the center 45 of the image 21a. The subject candidate with the shortest distance between the midpoint 41 of the shoulders and the center 45 of the image 21a is determined to be subject 10.

[0030] Furthermore, if there are two or more candidate subjects whose midpoint 41 of both shoulders is equal to the center 45 of image 21a, the subject skeleton extraction unit 25 further calculates the angle θ that the line connecting the shoulders of the candidate subjects makes with the horizontal direction of image 21a (parallel to the short side direction of the top plate 80) as a feature, and determines that the candidate subject whose angle θ of the line connecting the shoulders is close to zero, that is, close to the horizontal direction of image 21a, is subject 10.

[0031] Furthermore, the subject skeleton extraction unit 25 may determine that there is no subject 10 if the distance between the midpoint 41 of both shoulders of the skeleton identified as subject 10 and the center 45 of image 21a differs by a predetermined value or more from the average distance between the midpoint 41 of both shoulders of subject 10 and the center 45 of image 21a in multiple images 21a taken by the camera 21 after the start of fluoroscopic image acquisition.

[0032] As shown in Figure 5, the region of interest selection unit 26 sets multiple regions of interest 52 in the image of the subject 10 based on the position of the subject's skeleton extracted by the skeleton detection unit 24.

[0033] Meanwhile, the optical flow calculation unit 27 calculates the amount of movement (optical flow) of each corresponding pixel between the most recent image 21a and an image 21a acquired by the camera 21 at a predetermined time before the timing of acquisition of the most recent image 21a (Figure 6(a)). A known method such as the Farneback method is used to calculate the optical flow.

[0034] For example, the optical flow calculation unit 27 calculates the amount of movement (optical flow) of corresponding pixels between the most recently acquired image 21a and the image 21a acquired and rotated at a predetermined time before the acquisition timing, for each pixel.

[0035] The shadow detection unit 201 detects moving shadows contained in the most recent image 21a (Figure 6(b)). For example, it detects moving shadows in image 21a using foreground / background segmentation based on a mixture of normal distributions. As a foreground / background segmentation algorithm, for example, "BackgroundSubtractorMOG2" is used.

[0036] During the procedure, as shown in Figure 3, many medical personnel are around the subject 10, operating the endoscope or moving the endoscope itself. Therefore, image 21a may contain shadows that move in conjunction with the movements of the medical personnel, and the optical flow calculation unit 27 may mistakenly detect the optical flow of the moving shadows as the movement of the subject 10.

[0037] Therefore, in this embodiment, the shadow detection unit 201 detects moving shadows. The region of interest selection unit 126 removes from the subject 10's body movement determination any region of interest 52 that contains a shadow optical flow of a threshold or higher, from among the multiple regions of interest 52 set by the region of interest selection unit 26.

[0038] The determination unit 28 determines the number of regions of interest in the image 21a of the region of interest 52 that remain after the region of interest selection unit 126 has removed them, and determines whether or not there is body movement of the subject 10 based on whether this number is greater than a threshold (for example, 3).

[0039] A notification unit 29 is connected to the motion detection unit 20. When the determination unit 28 determines that motion has occurred, the notification unit 29 notifies the user of this fact. For example, the notification unit 29 displays a predetermined notification dialog 161 that notifies the user of the occurrence of motion on the display screen of the connected camera image display monitor 15 (see Figure 8). The notification unit 29 may also display the amount of movement calculated by the optical flow calculation unit 27 numerically or graphically on the display screen of the camera image display monitor 15 (see Figure 9). Alternatively, both of these may be displayed (see Figure 8).

[0040] Furthermore, the notification unit 29 may be configured to notify of body movement by changing the color of the edge of the image 21a displayed on the camera image display monitor 15, as shown in Figures 10(a) and (b). For example, the notification unit 29 may display a green line on the edge of the image 21a if there is no body movement, and a red line on the edge of the image 21a if there is body movement.

[0041] The monitor displaying the notification is not limited to the camera image display monitor 15, but may also be the fluoroscopy image display monitor 14 or the endoscope image display monitor 16.

[0042] The notification unit 29 may also notify the user of body movement by sound or light.

[0043] The operation of each part of the motion detection unit 20 will be explained below using the flowchart in Figure 7.

[0044] Furthermore, the functions of each part (22-26) of the motion detection unit 20 can be implemented by software. In this case, the motion detection unit 20 is configured using a computer equipped with a processor such as a CPU (Central Processing Unit) or GPU (Graphics Processing Unit) and memory, and the CPU implements these functions by reading and executing a program stored in memory. It is also possible to configure some or all of the motion detection unit 20 using hardware. For example, a circuit design can be performed to implement the functions of each part using a custom IC such as an ASIC (Application Specific Integrated Circuit) or a programmable IC such as an FPGA (Field-Programmable Gate Array).

[0045] (Step S101) The image acquisition unit 22 acquires images 21a from the camera 21 at predetermined time intervals. The timing at which the image acquisition unit 22 starts acquiring images may be when it receives a start command from the operator, or when the display of the endoscopic image on the endoscopic image display monitor 16 begins.

[0046] (Step S102) The image rotation unit 23 rotates image 21a by 90 degrees in a predetermined direction. As a result, the predetermined end on the top plate 80 on which the head of the subject 10 is mounted is positioned at the top of image 21a.

[0047] (Step S103) The skeleton detection unit 24 extracts the skeletal feature points (both ears, nose, both eyes, both shoulders, both elbows, both wrists, both hips, both knees, both ankles, etc.) of all human figures in image 21a and detects their two-dimensional coordinates (see Figure 4). This extracts the skeletons of all human figures.

[0048] (Step S104) If no skeleton is detected in step S103, the process returns to step S101 and captures the next image 21a. If a skeleton is detected, the process proceeds to step S105.

[0049] (Step S105) The subject skeleton extraction unit 25 calculates predetermined feature quantities using the position coordinates of the feature points of the skeletons of all human figures detected by the skeleton detection unit 24, and extracts the skeleton of the subject 10 mounted on the top plate based on the calculated feature quantities.

[0050] Furthermore, the subject skeleton extraction unit 25 distinguishes the subject 10 from the extracted subject 10, recognizing that the other human figures are medical personnel around the tabletop 80. It extracts only the human figure of subject 10 and displays it on the camera image display monitor 15 as shown in Figures 1(b) and (c).

[0051] (Step S106) The region of interest selection unit 26 partially sets the region of interest 52 based on the positions of predetermined feature points among the skeletal feature points of the subject 10, as shown in Figure 5.

[0052] Specifically, the region of interest selection unit 26 sets a region of interest 52 of a predetermined size centered on the skeletal feature points of the subject 10 detected by the skeletal detection unit 24, specifically at both shoulders, both elbows, both wrists, both hips, both knees, and both ankles.

[0053] Furthermore, the region of interest selection unit 26 determines the position of the subject 10's head and sets a region of interest 52 of a predetermined size relative to the head position. For example, the region of interest is set so that a part of the contour of the subject 10's occipital region is located inside the region of interest 52. Any method can be used to determine the head position, but for example, the region of interest selection unit 26 finds a line that passes through the coordinates of the ear closer to the center of the subject 10's body in the y-axis direction (longitudinal direction of the tabletop 80) of the two ears of the subject 10 detected by the skeleton detection unit 24, and is parallel to the x-axis direction (short-axis direction of the tabletop 80). The region of interest selection unit 26 also finds a line that passes through the midpoint of the line connecting both shoulders and is parallel to the y-axis direction. The region of interest selection unit 26 can set the intersection of the two lines found above as the head position 77.

[0054] The size of the region of interest 52 will be a predetermined, constant size. For example, it will be slightly larger than the circumference of an adult male arm that is 170 cm tall, or a predetermined size that does not overlap with the operator's hand.

[0055] (Step S107) The optical flow calculation unit 27 calculates the amount of movement (optical flow) of corresponding pixels between the image 21a rotated in step S102 and the image 21a acquired and rotated at a predetermined time before the acquisition timing, pixel by pixel (see Figure 6(a)). The image 21a used to calculate the amount of movement is not an image in which only the person's image of the subject 10 is extracted, but the entire image including all of the person's image.

[0056] (Step S108) The determination unit 28 determines whether optical flow was detected in the area surrounding the skeleton of the subject extracted in step S107. If optical flow is detected in the area surrounding the skeleton of the subject, the process proceeds to step S109. If no optical flow is detected in the area surrounding the skeleton of the subject, the process returns to step S101.

[0057] (Step S109) The shadow detection unit 201 detects the region of the moving shadow contained in the image 21a (Figure 6(b)). The same image used in step S107 for calculating the optical flow is used.

[0058] (Step S110) The region of interest selection unit 126 determines whether there is a region of interest 52 in which the pixels of the shadowed region detected in step S109 are present in a proportion equal to or greater than a predetermined threshold. If there is a region of interest 52 in which the pixels of the shadowed region are present in a proportion equal to or greater than a predetermined threshold, the process proceeds to step S111. If there is no such region of interest 52, the process proceeds to step S112.

[0059] (Step S111) The region of interest selection unit 126 removes the region of interest 52 from the region of interest 52 if it determined in step S110 that the proportion of pixels in the shadow region is above a predetermined threshold (see Figure 6(c)).

[0060] (Step S112) The determination unit 28 determines the regions of interest 52 remaining after the processing in step S111, where the number of pixels whose pixel movement exceeds a threshold is greater than a predetermined value (for example, 20% of the number of pixels in one region of interest 52).

[0061] Furthermore, the determination unit 28 determines the number of regions of interest in which the amount of movement is greater than a predetermined value. If this number is greater than a predetermined value (for example, 3), it determines that the optical flow of the region of interest 52 exceeds the threshold, and that body movement has occurred in the subject 10, and proceeds to step S113. If the number of regions of interest in which the amount of movement is greater than a predetermined value is less than or equal to the threshold, there is no body movement, and the process proceeds to step S114.

[0062] (Step S113) If the determination unit 28 determines in step S112 that body movement has occurred, the notification unit 29 notifies the user that body movement has occurred.

[0063] Specifically, as shown in Figure 8, the notification unit 29 displays a pre-defined notification dialog 161 on the camera image display monitor 15 to notify the user of body movement. Alternatively, as shown in Figure 9, the notification unit 29 displays the amount of movement calculated by the optical flow calculation unit 27 as a numerical value or graph 162 to notify the user of body movement. Alternatively, the notification unit 29 displays both the notification dialog 161 and the graph 162. Furthermore, as shown in Figure 10, the notification unit 29 may also notify the user of body movement by changing the color of the border of the image 21a displayed on the camera image display monitor 15.

[0064] Furthermore, the notification unit 29 may emit an alarm sound or light if it detects body movement.

[0065] (Steps S115, S116) The notification unit 29 determines whether a predetermined time (for example, 10 seconds) has elapsed since the start of the body motion notification in step S113 (step S115). If the predetermined time has elapsed, it proceeds to step S116 to stop the notification and returns to step S101.

[0066] On the other hand, if a predetermined amount of time has not elapsed since the start of the body movement notification in step S115, the system returns to step S101 without stopping the notification.

[0067] (Step S114) On the other hand, if the result is NO in step S110, or if the result is NO in step S112, proceed to step S114 to determine if a body movement notification is in progress. If a body movement notification is in progress, proceed to step S115. On the other hand, if a body movement notification is not in progress, return to step S101.

[0068] By performing steps S114 to S116 in this manner, it is possible to continue the body movement notification for a predetermined period of time (for example, 10 seconds).

[0069] As described above, according to Embodiment 1, an image of the subject 10 can be extracted and displayed, and optical flow of moving shadows of medical personnel, etc., included in the image can be removed, and it can be determined whether or not there is body movement in the subject 10, and the user can be notified.

[0070] Therefore, according to Embodiment 1, it is possible to avoid misjudging body movement due to moving shadows.

[0071] Therefore, the X-ray imaging device of this embodiment can accurately detect the body movements of the subject while fluoroscopic images are being taken, allowing the physician to easily understand the patient's body movements even during a procedure.

[0072] <<Embodiment 2>> In embodiments 2-1, 2-2, and 2-3, if a protective curtain 210 is installed around the X-ray irradiation device 11 as shown in Figures 11(a) and (b) to prevent unnecessary exposure of the subject 10, the image of the protective curtain 210 is included in the image of the camera 21, which may cause the skeleton detection unit 24 to erroneously detect the joints of the subject 10's skeleton. For example, in the example in Figure 11(a), the joint position of the arm of the medical worker 220 is erroneously detected, and the region of interest 52a is set accordingly.

[0073] Therefore, embodiments 2-1, 2-2, and 2-3 are equipped with a designation unit that receives a designation from the operator as to whether or not a protective curtain is being used, or a detection unit that detects a protective curtain, and exclude a predetermined region of interest 52a from the region of interest 52 used by the region of interest selection unit 126 for determining body movement (Figure 11(b). This will be explained in detail below.

[0074] <Embodiment 2-1> As shown in Figure 12, the X-ray imaging apparatus 1 of Embodiment 2-1 is equipped with a protective curtain button 202, which is a designation unit that receives a designation from the operator indicating whether or not a protective curtain is being used.

[0075] (Steps S101~S108) As shown in the flow chart in Figure 13, steps S101 to S108 are performed in the same manner as in Embodiment 1.

[0076] (Step S120) Next, the region of interest selection unit 126 proceeds to step S120, where it determines whether the operator has turned on the protective curtain button 202. If the protective curtain button 202 is turned on, the region of interest selection unit 126 proceeds to step S121.

[0077] If the protective curtain button 202 is OFF, proceed to step S112.

[0078] (Step S121) In step S121, the region of interest selection unit 126 excludes the other regions of interest 52a from the regions of interest, keeping only the predetermined regions of interest (in this case, the regions of interest of the head and shoulders) from the regions of interest 52 set by the region of interest selection unit 26 in step S106.

[0079] (Step S112) The determination unit 28 proceeds to step S112 and, using only the region of interest (in this case, the region of interest of the head and shoulders) 52 remaining from step S121, determines whether the optical flow exceeds a threshold, thereby determining whether body movement has occurred, similar to Embodiment 1.

[0080] (Steps S113-S116) Similar to Embodiment 1, the occurrence of body movement is notified.

[0081] As a result, when the protective curtain button 202 is turned on, as shown in Figure 14(b), the region of interest that was mistakenly set to the joint positions of the medical professional 220 can be excluded, and the occurrence of body movement can be determined using only the region of interest 52 of the subject's head and shoulders, thereby avoiding misjudgment of the occurrence of body movement.

[0082] On the other hand, when the protective curtain button 202 is off, as shown in Figure 14(a), in step S121, the determination unit 28 uses all regions of interest 52 to determine the occurrence of body movement. Therefore, the body movement of the subject 10 can be determined with high accuracy.

[0083] <Embodiment 2-2> The X-ray imaging apparatus 1 of Embodiment 2-2 is equipped with a protective curtain detection unit 203 that detects whether or not a protective curtain 210 is attached to the X-ray irradiation apparatus 11, as shown in Figure 15.

[0084] For example, recesses are provided around the X-ray irradiation device 11 for hooking the protective curtain 210. The protective curtain detection unit 203 is a pressure sensor located at the position of the recess. When the protective curtain detection unit 203 detects the weight of the protective curtain 210 as pressure, the protective curtain 210 is attached to the X-ray irradiation device 11.

[0085] The X-ray imaging apparatus 1 of Embodiment 2-2 performs step S130 instead of step S120 of Embodiment 2-1, as shown in the flowchart in Figure 16.

[0086] (Step S130) In step S130, the region of interest selection unit 126 determines whether the protective curtain detection unit 203 has detected that the protective curtain 210 is draped over the X-ray irradiation device 11. If the protective curtain 210 is draped, the region of interest selection unit 126 proceeds to step S121 and excludes the other regions of interest 52a from the region of interest, leaving only the predetermined region of interest (in this case, the region of interest of the head and shoulders).

[0087] On the other hand, if the protective curtain 210 is not detected, the region of interest selection unit 126 proceeds to step S112.

[0088] Other configurations and operations are the same as in Embodiment 2-1, so their description will be omitted.

[0089] <Embodiment 2-3> In the case of X-ray fluoroscopy, in addition to attaching a protective curtain 210 to the X-ray irradiation device 11, a towel or blanket may also be placed over the subject 10.

[0090] Therefore, in Embodiment 2-3, a region of interest range selection button 204 is placed on the X-ray imaging device 1 as shown in Figure 17 to receive a specification from the operator for the area not covered by the protective curtain or towel blanket, that is, the range of the region of interest used for determining body movement.

[0091] Specifically, as shown in Figures 18(a), (b), and (c), for example, the region of interest selection button 204 includes a button 204a for selecting the entire region of interest 52 of the subject 10, a button 204b for selecting only the region of interest 52 of the upper body when the lower body of the subject 10 is covered with a towel blanket, and a button 204c for selecting only the region of interest 52 of the head and shoulders when a protective curtain 210 is used.

[0092] The X-ray imaging apparatus 1 of Embodiment 2-3 performs steps S140 and S141 instead of steps S130 and S121 of Embodiment 2-2, as shown in the flowchart in Figure 19.

[0093] (Step S140) In step S140, the region of interest selection unit 126 determines which of the region of interest range selection buttons 204 is turned on.

[0094] (Step S141) In step S140, the region of interest selection unit 126 excludes unselected regions of interest according to the region of interest range selection button 204 that is turned on.

[0095] Other configurations and operations are the same as in Embodiment 2-2, so their description will be omitted.

[0096] In Embodiment 2-3, by pre-providing multiple desired types of area of ​​interest selection buttons 204, it becomes possible to set multiple types of area of ​​interest selection ranges, such as when using a towel blanket.

[0097] <<Embodiment 3>> If the subject 10 moves during X-ray fluoroscopy, a medical professional may restrain the subject 10 to stop its movement. In such cases, the image of the subject 10 (skeleton) and the image of the medical professional (skeleton) may come close to or intersect in the image captured by the camera 21. In this case, the judgment unit 28 may misidentify the movement of the medical professional as the movement of the subject 10.

[0098] In embodiments 3-1, 3-2, and 3-3, if the image of the subject 10 and the image of the medical professional overlap on the image, the region of interest selection unit 126 excludes all regions of interest 52 and does not use them for determining body movement.

[0099] <Embodiment 3-1> In Embodiment 3-1, as shown in Figure 20, a large second region of interest 152 is set to surround the subject 10, and if the skeleton other than the subject 10 is included in the second region of interest 152, that image is excluded from the body movement detection.

[0100] The X-ray imaging apparatus 1 of Embodiment 3-1 performs steps S150 and S151 instead of steps S109 to S111 of Embodiment 1, as shown in the flowchart in Figure 21.

[0101] (Steps S101~S108) Steps S101 to S108 are performed in the same manner as in Embodiment 1. In step S106, the region of interest selection unit 26 sets the region of interest 52 to feature points such as joints of the subject 10.

[0102] (Step S150) In step S150, the region of interest selection unit 26 sets a large second region of interest 152 surrounding the subject 10.

[0103] Specifically, in step S105, the region of interest selection unit 26 determines the coordinates of the left, right, top, and bottom edges of the rectangular region containing the subject 10 based on the coordinates of the feature points of the subject 10 calculated, and sets the second region of interest 152. For example, if the horizontal direction of the image 21a rotated in step S102 is the x-axis and the vertical direction is the y-axis, then the left edge of the region is set to a predetermined distance (e.g., 30 pixels) to the left of the leftmost feature point in the x-axis direction among the feature points (both ears, nose, both eyes, both shoulders, both elbows, both wrists, both hips, both knees, both ankles), the right edge of the region is set to a predetermined distance (e.g., 30 pixels) to the right of the rightmost feature point in the x-axis direction, and the bottom edge of the region is set to a predetermined distance (e.g., 30 pixels) below the bottommost feature point in the y-axis direction. Furthermore, if we let Ye be the y-coordinate of the ear located lower in the y-axis direction, and Ys be the y-coordinate of the shoulder located upper in the y-axis direction, then the upper boundary of the region is defined as Ye-(Ys-Ye). This allows us to define a large second region of interest 152 surrounding the subject 10.

[0104] (Step S151) In step S151, the region of interest selection unit 126 determines whether any skeletons other than those of the subject 10 extracted in step S105 are located inside the second region of interest 152. If any skeletons other than those of the subject 10 are located inside the second region of interest 152, the region of interest selection unit 126 excludes all regions of interest 52 and returns to step S101. As a result, the determination unit 28 does not use that image for body movement determination.

[0105] On the other hand, if the region of interest selection unit 126 determines that no skeleton other than the skeleton of the subject 10 is inside the second region of interest 152, the process proceeds to step S112. The determination unit 28 uses all regions of interest 52 to determine whether or not body movement has occurred.

[0106] Other configurations and operations are the same as in Embodiment 1, so their description will be omitted.

[0107] <Embodiment 3-2> In Embodiment 3-1, the region of interest selection unit 126 was configured to exclude all regions of interest 52 if skeletons other than the subject 10 were located in the second region of interest 152, and not use those images for body motion detection. However, in Embodiment 3-1, as shown in Figure 22(a), skeletons other than the subject 10 are located in the second region of interest 152, but even if the skeletons other than the subject 10 do not overlap with the subject 10, their images are not used for body motion detection.

[0108] Therefore, in Embodiment 3-2, as shown in Figure 22(b), when the skeletal lines connecting the joints of the subject 10 intersect with skeletal lines other than those of the subject 10, the region of interest selection unit 126 excludes all regions of interest 52 and does not use those images for body movement determination.

[0109] The X-ray imaging apparatus 1 of Embodiment 3-2 performs steps S160 and S161 instead of steps S150 and S151 of Embodiment 3-1, as shown in the flowchart in Figure 23.

[0110] (Steps S101~S108) Steps S101 to S108 are performed in the same manner as in Embodiment 1. In step S103, all skeletons are detected, and in step S105, the skeleton of the subject 10 is extracted.

[0111] (Step S160) In step S160, the region of interest selection unit 126 calculates all skeletal lines of the subject 10 and all skeletal lines of the other skeletons from the joint positions of the subject and the other skeletons calculated in step S103.

[0112] (Step S161) In step S161, the region of interest selection unit 126 determines whether there is an intersection between the skeletal line of the subject 10 and the skeletal line of a person other than the subject 10.

[0113] As shown in Figure 22(b), if the region of interest selection unit 126 finds that the skeletal lines of the subject 10 intersect with the skeletal lines of other subjects 10, it excludes all regions of interest 52, returns to step S101, and does not use that image for body motion detection.

[0114] On the other hand, if the skeletal lines of subject 10 do not intersect with the skeletal lines of subjects other than subject 10, the region of interest selection unit 126 proceeds to step S112. The determination unit 28 uses the region of interest 52 to determine whether or not body movement has occurred.

[0115] Other configurations and operations are the same as in Embodiment 3-1, so their description is omitted.

[0116] <Embodiment 3-3> In embodiments 3-1 and 3-2, if a skeleton other than the subject 10 is located in the second region of interest 152, or if the skeleton of the subject 10 intersects with a skeleton other than the subject 10, the region of interest selection unit 126 excludes all regions of interest 52 and does not use the entire image for motion detection. Therefore, the detection results of the optical flow in the region of interest of the subject 10 that does not overlap with a skeleton other than the subject 10 cannot be used.

[0117] Therefore, in Embodiment 3-3, as shown in Figure 24, if a skeletal line other than that of the subject 10 passes through the region of interest 52, the region of interest selection unit 126 excludes only that region of interest 52 from the determination of body movement.

[0118] The X-ray imaging apparatus 1 of Embodiment 3-3 performs steps S170 to S172 instead of steps S160 and S161 of Embodiment 3-2, as shown in the flowchart in Figure 25.

[0119] (Steps S101~S108) Steps S101 to S108 are performed in the same manner as in Embodiment 1. In step S103, all skeletons are detected, and in step S105, the skeleton of the subject 10 is extracted. In step S106, a region of interest 52 is set for feature points such as joints of the subject 10.

[0120] (Step S170) In step S170, the region of interest selection unit 126 calculates all skeletal lines other than the subject 10 from the joint positions of the subject and other skeletons calculated in step S103.

[0121] (Step S171) In step S171, the region of interest selection unit 126 determines whether the skeletal lines other than those of the subject 10 pass through the region of interest 52 of the subject 10. If the skeletal lines other than those of the subject 10 pass through the region of interest 52, the region of interest selection unit 126 proceeds to step S172. If they do not pass through, the process proceeds to step S112.

[0122] (Step S172) In step S172, the region of interest selection unit 126 excludes the region of interest 52 through which skeletal lines other than those of the subject 10 pass, and proceeds to step S112.

[0123] In step S112, the determination unit 28 determines whether or not the subject 10 is moving using the optical flow of the region of interest that was not excluded.

[0124] This makes it possible to utilize the detection results of the optical flow in the region of interest 52 of subject 10 that does not overlap with the skeleton other than subject 10.

[0125] Other configurations and operations are the same as in Embodiment 3-1, so their description is omitted.

[0126] <<Embodiment 4>> In Embodiments 2 and 3, when the protective curtain 210 was hung over the X-ray irradiation device 11, the region of interest 52 of the subject 10's head and shoulders was used. However, as shown in Figure 26(a), if the protective curtain 210 covers the subject 10's shoulders, the skeleton detection unit 24 is unable to extract the subject 10's skeleton, and therefore cannot set a region of interest 52 for the head, which is not covered by the protective curtain 210. As a result, the determination unit 28 cannot detect body movement.

[0127] Therefore, the X-ray imaging apparatus 1 of Embodiment 4 is equipped with a region of interest storage unit 180, as shown in Figure 27. The region of interest selection unit 126 stores the position of the region of interest 52 set by the region of interest selection unit 26 in the region of interest storage unit 180 for each image. If the shoulders of the subject 10 are covered by a protective curtain 210 or the like and the region of interest 52 cannot be set, the region of interest selection unit 126 reads the position of the region of interest set for the previous image and instructs the region of interest selection unit 26 to set the region of interest 52 at the same position in the current image (Figure 26(b)).

[0128] In Embodiment 4, the X-ray imaging apparatus 1 does not perform steps S109 to S111 of Embodiment 1, as shown in the flowchart in Figure 28. If the skeleton could not be detected in step S104, the region of interest 52 is set to the previously saved position in steps S191 and S192.

[0129] (Steps S101-S103, S105) Steps S101 to S103 and S105 are performed in the same manner as in Embodiment 1. In step S103, the skeleton detection unit 24 detects all skeletons, and the process proceeds to step S105, where the subject skeleton extraction unit 25 extracts the skeleton of the subject 10.

[0130] (Steps S190, S106) In step S190, the subject skeleton extraction unit 25 determines whether the subject skeleton has been extracted. If it has been extracted, the process proceeds to step S106, where the region of interest selection unit 26 sets the region of interest to feature points such as joints.

[0131] (Step S193) The region of interest selection unit 126 proceeds to step S193, where it stores the position where the region of interest 52 was set in step S106 in the region of interest storage unit 180.

[0132] (Step S191) On the other hand, if the skeleton detection unit 24 determines in step S190 that it has not detected a skeleton, the process proceeds to step S191, and the skeleton detection unit 24 notifies the operator that a skeleton has not been detected. For example, the camera image display monitor 15 displays that the region of interest 52 cannot be set because a skeleton has not been detected. It also displays that the region of interest 52 should be set to the same value as the previous image.

[0133] (Step S192) Proceeding to step S192, the region of interest selection unit 126 reads the setting position of the region of interest 52 of the previous image stored in the region of interest storage unit 180 and instructs the region of interest selection unit 26 to set the region of interest 52 to the same position in the current image. The region of interest selection unit 26 sets the region of interest 52 to the same position as in the previous image.

[0134] (Steps S107, S108, S112~S116) The process proceeds to step S107, where the optical flow calculation unit 27 calculates the optical flow for the current image.

[0135] The determination unit 28 proceeds to steps S108 and S109, where it determines whether the optical flow exceeds a threshold for the region of interest set in the current image in step S106 or S192, and determines whether or not there is body movement.

[0136] Steps S113 to S116 are the same as in Embodiment 1, so their explanation will be omitted.

[0137] According to Embodiment 4, even if the protective curtain 210 or the like covers the shoulders of the subject 10 and the skeleton detection unit 24 cannot extract the skeleton of the subject 10, it can still set a region of interest 52 for the head and other parts not covered by the protective curtain 210, thereby enabling the detection of body movement. [Explanation of Symbols]

[0138] 1. X-ray imaging device 10 subjects 11 X-ray irradiation device 12 X-ray detectors 13 X-ray image generation section 14. Transparency Image Display Monitor 15 Camera Image Display Monitor 16 Endoscopic image display monitor 20 Motion detection unit 21 Cameras 21a Image 22 Image acquisition unit 23 Image rotation section 24 Skeleton detection unit 25 Subject skeleton extraction section 26. Area of ​​Interest Selection Section 27 Optical Flow Calculation Unit 28 Judgment section 29 Notification Department 41 Midpoint 45 center 50 X-ray support 50a Longitudinal direction 50b Width direction 50c slope 51 Rectangular area 52 Areas of Interest 52a Area of ​​Interest 60 X-ray vertical movement mechanism 61 X-ray tilt mechanism 62 Pressing tool 70 Stands 77 position 80 Top plate 81 Top panel support section 152 Second Area of ​​Interest 161 Notification Dialog 162 Graphs 180 Area of ​​Interest Memory 201 Shadow detection unit 202 buttons 203 Protective Curtain Detection Unit 204 Select area of ​​interest button 204a button 204b button 204c button 210 Protective Curtain 220 healthcare workers

Claims

1. The device comprises a top plate on which a subject is mounted, an X-ray irradiation device that irradiates the subject with X-rays, an X-ray detector that detects the X-rays irradiated from the X-ray irradiation device and transmitted through the subject, and a motion detection unit that receives an image from a camera that images the subject, processes the image, and detects the occurrence of motion of the subject. The aforementioned motion detection unit includes a skeleton detection unit, an optical flow calculation unit, a region of interest selection unit, and a determination unit. The skeleton detection unit detects the skeleton of the human figure contained in the image, and extracts the skeleton of the subject mounted on the top plate based on the detected skeleton. The optical flow calculation unit calculates the amount of movement of corresponding pixels between the image and an image acquired by the camera at a predetermined time before the timing at which the image was acquired. The region of interest selection unit sets multiple regions of interest in the image of the subject based on the position of the subject's skeleton extracted by the skeleton detection unit. The determination unit determines whether body movement is occurring in the subject based on the amount of movement of the subject's pixels within the region of interest. The aforementioned region of interest selection unit includes a region of interest selection unit, The region of interest selection unit either selects one or more regions of interest from among a plurality of regions of interest set by the region of interest selection unit to be used by the determination unit for determining body movement, or specifies to the region of interest selection unit the location of the region of interest to be used by the determination unit for determining body movement. An X-ray imaging apparatus characterized by the following features.

2. An X-ray imaging apparatus according to claim 1, The motion detection unit includes a shadow detection unit, The shadow detection unit detects pixels of moving shadows included in the image, The region of interest selection unit removes from among the plurality of regions of interest set by the region of interest selection unit any region of interest in which the shadow pixels are present in a predetermined value or more. The X-ray imaging apparatus is characterized in that the determination unit determines whether or not body movement is occurring in the subject based on the amount of movement of the subject's pixels in the region of interest that remain after the region of interest selection unit has removed them.

3. An X-ray imaging apparatus according to claim 1, The aforementioned motion detection unit includes a protective curtain designation / detection unit. The protective curtain designation / detection unit receives from the operator whether or not a protective curtain is present, or the range of the region of interest to be used for determining body movement, or it detects whether or not a protective curtain is present. The interest region selection unit removes a predetermined interest region from among the multiple interest regions set by the interest region selection unit, in accordance with the result received by the protective curtain designation / detection unit or the detection result. The X-ray imaging apparatus is characterized in that the determination unit determines whether or not body movement is occurring in the subject based on the amount of movement of the subject's pixels in the region of interest that remain after the region of interest selection unit has removed them.

4. An X-ray imaging apparatus according to claim 3, wherein the protective curtain designation / detection unit is a button that accepts the presence or absence of a protective curtain, or a button that allows selection of multiple ranges of the region of interest used for determining body movement.

5. An X-ray imaging apparatus according to claim 3, characterized in that the protective curtain designation / detection unit is a sensor provided at the location where the protective curtain of the X-ray irradiation apparatus is attached.

6. An X-ray imaging apparatus according to claim 1, The region of interest selection unit determines in the image whether any skeletons other than the subject's skeleton are approaching or intersecting the subject's skeleton, and if any skeletons other than the subject's skeleton are approaching or intersecting, removes some or all of the multiple regions of interest set by the region of interest selection unit. The determination unit determines whether body movement is occurring in the subject based on the amount of movement of the subject's pixels in the region of interest that remain after the region of interest selection unit has removed them. An X-ray imaging apparatus characterized by the following features.

7. The X-ray imaging apparatus according to claim 6, wherein the region of interest selection unit further sets a second region of interest that surrounds the entire subject, The X-ray imaging apparatus is characterized in that the region of interest selection unit removes the entire region of interest if the second region of interest includes skeletons other than the subject, and the determination unit does not determine whether body movement has occurred in the image.

8. An X-ray imaging apparatus according to claim 6, wherein the region of interest selection unit determines whether there is an intersection between the skeletal lines of a subject other than the subject and the skeletal lines of the subject, and if there is an intersection, the entire region of interest is removed, and the determination unit does not determine whether body movement has occurred in the image.

9. An X-ray imaging apparatus according to claim 6, wherein the region of interest selection unit determines whether skeletal lines other than those of the subject pass through the region of interest of the subject, and if so, removes the region of interest through which the skeletal lines other than those of the subject pass from a plurality of regions of interest set by the region of interest selection unit, and the determination unit determines the occurrence of body motion using the remaining regions of interest.

10. An X-ray imaging apparatus according to claim 1, The motion detection unit includes a storage unit, The aforementioned region of interest selection unit stores the location of the set region of interest in the storage unit. If the skeleton detection unit cannot detect a skeleton in the image, the region of interest selection unit instructs the region of interest selection unit to set the region of interest to the position of the region of interest stored in the storage unit for previous images. The aforementioned region of interest selection unit sets the region of interest at the position read from the storage unit, The determination unit determines the occurrence of body movement using the region of interest set at the position read from the storage unit. An X-ray imaging apparatus characterized by the following features.

11. A method for processing images captured by a camera in an X-ray imaging apparatus equipped with a camera, The skeleton of the human figure contained in the aforementioned image is detected, and based on the detected skeleton, the skeleton of the subject mounted on the top plate of the X-ray imaging device is extracted. The amount of movement of the corresponding pixels is calculated between the aforementioned image and an image acquired by the camera at a predetermined time before the time the aforementioned image was acquired. Based on the position of the subject's skeleton, multiple regions of interest are set in the image of the subject, and one or more regions of interest to be used for motion detection are selected from the set regions of interest, or the location of one or more regions of interest to be used for motion detection is specified, and a region of interest is set at that location. Using the selected region of interest or the amount of movement of pixels in the region of interest set by specifying a location, it is determined whether or not body movement is occurring in the subject. An image processing method for an X-ray imaging apparatus equipped with a camera, characterized by the following.

Images