JP2024034220A - X-ray computed tomography apparatus and estimation method - Google Patents

Abstract

To more accurately position a subject.SOLUTION: An X-ray computed tomography apparatus according to the embodiment includes an image acquisition section, a position acquisition section, and a position estimation section. The image acquisition section acquires a first image in which a subject is captured from a first direction and a second image in which the subject is captured from a second direction that is different from the first direction. The position acquisition section acquires first position information corresponding to the positions of the plurality of different parts of the subject on the basis of the first image, and acquires second position information corresponding to the positions of the plurality of different parts of the subject on the basis of the second image. The position estimation section estimates the position of the subject on a bed on which the subject is placed, on the basis of the first position information and the second position information.SELECTED DRAWING: Figure 1

Description

Embodiments disclosed in this specification and the drawings relate to an X-ray computed tomography apparatus and an estimation method.

2. Description of the Related Art Conventionally, in the medical field, doctors make diagnoses using medical images taken by various medical image diagnostic devices (also called modalities). Here, examples of the medical image diagnostic apparatus include an X-ray computed tomography apparatus (hereinafter referred to as an X-ray CT (Computed Tomography) apparatus), a photoacoustic imaging apparatus (hereinafter referred to as a PAT (PhotoAcoustic Tomography) apparatus), Examples include a magnetic resonance imaging device (hereinafter referred to as an MRI (Magnetic Resonance Imaging) device).

For example, with regard to medical image diagnostic equipment such as an X-ray CT device, by recognizing the body part of the subject from images obtained by imaging the subject with two cameras, the bed is controlled and the subject is moved to the target position. Techniques for alignment are known.

However, the object to be examined, such as the human body, has an uneven shape, and the distance between the camera and the region varies depending on the region, so positioning is performed individually using information about the region obtained from images captured by each camera. However, accurate positioning may not be possible.

Patent No. 4484462 Japanese Patent Application Publication No. 2006-187422

One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to perform more accurate positioning of a subject. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problems. Problems corresponding to each effect of each configuration shown in the embodiments described later can also be positioned as other problems.

The X-ray CT apparatus according to the embodiment includes an image acquisition section, a position acquisition section, and a position estimation section. The image acquisition unit acquires a first image of the subject taken from a first direction, and a second image taken of the subject from a second direction that is different from the first direction. . The position acquisition unit acquires first position information corresponding to the positions of the plurality of different parts of the subject based on the first image, and acquires first position information corresponding to the positions of the plurality of different parts based on the second image. second position information corresponding to the position of is acquired. The position estimation unit estimates the position of the subject on the bed on which the subject is placed, based on the first position information and the second position information.

FIG. 1 is a diagram showing an example of the configuration of an X-ray CT apparatus according to the first embodiment. FIG. 2 is a flowchart showing the flow of object alignment processing performed by the X-ray CT apparatus according to the first embodiment. FIG. 3 is a flowchart showing the flow of automatic alignment processing in step S103 shown in FIG. FIG. 4 is a diagram illustrating an example of a first image acquired by the image acquisition unit according to the first embodiment. FIG. 5 is a flowchart showing the flow of the first location information acquisition process in step S201 shown in FIG. FIG. 6 is a diagram illustrating an example of acquisition of first position information performed by the position acquisition unit according to the first embodiment. FIG. 7 is a diagram illustrating an example of a second image acquired by the image acquisition unit according to the first embodiment. FIG. 8 is a flowchart showing the flow of the second location information acquisition process in step 203 shown in FIG. FIG. 9 is a diagram illustrating an example of acquisition of second position information performed by the position acquisition unit according to the first embodiment. FIG. 10 is a diagram illustrating an example of estimating the position of the subject in the Y-axis direction performed by the position estimation unit according to the first embodiment. FIG. 11 is a diagram illustrating an example of estimating the position of the subject in the X-axis direction and the Z-axis direction, which is performed by the position estimation unit according to the first embodiment. FIG. 12 is a flowchart showing the flow of automatic alignment processing according to the second embodiment.

Hereinafter, embodiments of an X-ray CT apparatus and an estimation method will be described in detail with reference to the drawings.

(First embodiment)
FIG. 1 is a diagram showing an example of the configuration of an X-ray CT apparatus 1 according to the first embodiment.

As shown in FIG. 1, the X-ray CT apparatus 1 includes a gantry device 10, a bed device 30, and a console device 40. Hereinafter, the longitudinal direction of the top plate 32 of the bed device 30 will be defined as the Z-axis direction, the lateral direction will be defined as the X-axis direction, and the direction perpendicular to the floor surface will be defined as the Y-axis direction.

The gantry device 10 includes a rotation section 15 (not shown), a rotation drive section 16 (not shown), a gantry/bed control section 17, a camera 20, a camera 21, and a camera control section 22.

The rotating section 15 includes an X-ray tube 11 , an X-ray tube control section 12 , an X-ray detector 13 , and a data acquisition section 14 . The rotating portion 15 has an annular shape and rotates around the center of the annular ring. Inside the rotating section 15, an X-ray tube 11 and an X-ray detector 13 are arranged to face each other. The rotation drive section 16 rotationally drives the rotation section 15 .

The X-ray tube 11 emits X-rays from an X-ray focal point, and emits the X-rays toward the subject 60 placed thereon. The X-ray tube control section 12 includes a high voltage generation circuit, and supplies high voltage power to the X-ray tube 11 to cause it to emit X-rays. By controlling the high voltage generation circuit, the quality and dose of X-rays emitted from the X-ray tube 11 can be controlled.

The X-ray detector 13 electrically detects the X-rays emitted by the X-ray tube 11 and passed through the subject 60, and outputs an electrical signal corresponding to the intensity of the X-rays to the data acquisition unit 14.

The data collection unit 14 collects electrical signals output from the X-ray detector 13 as digital data. The data collection unit 14 is connected to the control bus 41 of the console device 40, and the collected digital data is transmitted to the console device 40.

The gantry/bed control unit 17 is connected to the control bus 41 of the console device 40, and controls the X-ray tube control unit 12, rotation drive unit 16, and bed drive unit 33 according to instructions from the scan control unit 50. , enables scanning of the subject 60. Note that the gantry/bed control section 17 may be included in the gantry device 10 or the console device 40.

The camera 20 and the camera 21 include an imaging optical system, an image sensor, a CPU (Central Processing Unit), an image processing circuit, a ROM (Read Only Memory), a RAM (Random Access Memory), at least one communication I/F (Interface), etc. , has a configuration as a general camera. An image is captured by focusing a light beam from a subject onto an image pickup element such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor) sensor by an imaging optical system including an optical element such as a lens. The imaging optical system includes a lens group, and each camera also includes a lens drive control section that controls zoom and focus by driving the lens group in the optical axis direction. The electrical signal output from the image sensor is converted into digital image data by an A/D (Analog to Digital) converter, subjected to various image processing by an image processing circuit, and then output to an external device. Note that the image processing performed by the image processing circuit may be performed by a processing section of the external device after the digital image data is output to the external device via the communication I/F.

In this embodiment, the cameras 20 and 21 use image sensors that mainly receive light in the visible light range and take images. Note that the examples of the cameras 20 and 21 are not limited to this, and may be cameras that receive light in the infrared region to take images, or cameras that receive light in multiple wavelength regions such as visible light and infrared light. It may also include a camera that receives light and takes an image. Hereinafter, the images captured by the cameras 20 and 21 will be referred to as optical images. Here, the camera 20 and the camera 21 may be installed at a location other than the gantry device 10, and only need to be installed at a position where the subject 60 can be photographed. For example, as shown in FIG. 1, in this embodiment, the camera 20 is installed above the gantry device 10 so as to image the subject 60 from diagonally above. Further, the camera 21 is installed on the inward side of the gantry device 10 so as to image the subject 60 from the side.

The camera control unit 22 is connected to the control bus 41 of the console device 40 and controls the cameras 20 and 21 according to instructions from the image acquisition unit 52. The camera control unit 22 includes at least one communication I/F for supplying power to the cameras 20 and 21, transmitting and receiving control signals, transmitting and receiving image signals, and the like. Note that the camera 20 and the camera 21 may be provided with a power source for driving them alone without receiving power supply from the gantry device 10. The camera control unit 22 also controls various imaging parameters such as zoom, focus, and aperture value of the cameras 20 and 21 by transmitting control signals to the cameras 20 and 21 via the communication I/F. Can be done. Note that the camera 20 and the camera 21 may be provided with a pan head that can automatically pan and tilt, and also receive a pan and tilt control signal, so that the position and orientation can be controlled by pan and tilt driving. In addition, the camera 20 and the camera 21 are provided with a drive unit and a drive control unit for electrically controlling the position and orientation of the camera 20 and the camera 21, and the position and orientation of the camera 20 and the camera 21 are controlled based on a control signal from the camera control unit 22. Good too. Note that the camera control unit 22 may be included in the gantry device 10 or the console device 40.

The bed device 30 includes a base 31, a top plate 32, and a bed drive section 33. The bed device 30 moves the base 31 up and down in the Y-axis direction, moves the top plate 32 back and forth in the Z-axis direction, and moves it left and right in the X-axis direction by the bed drive unit 33. The mounted subject 60 can be inserted into the gantry device 10. Here, the bed device 30, the base 31, and the top plate 32 are examples of a bed.

The console device 40 includes a control bus 41 , a processing circuit 42 , a memory 43 , a nonvolatile memory 44 , a communication I/F 45 , an input I/F 46 , and a display 47 . The processing circuit 42, memory 43, nonvolatile memory 44, communication I/F 45, input I/F 46, and display 47 are connected to the control bus 41 and can exchange data with each other via the control bus 41. It is configured as follows.

The processing circuit 42 controls each part of the console device 40. For example, the processing circuit 42 controls each part of the console device 40 according to a program stored in the nonvolatile memory 44, using the memory 43 as a work memory.

The memory 43 stores various information necessary for processing performed by each part of the console device 40. For example, the memory 43 is configured with a RAM or the like (volatile memory using a semiconductor element or the like).

The nonvolatile memory 44 stores tomographic images obtained from projection data transmitted from the data collection unit 14 of the gantry device 10, optical images transmitted from the cameras 20 and 21, and various programs for operating the processing circuit 42. etc. to be memorized. The nonvolatile memory 44 is composed of, for example, an HDD (Hard Disk Drive), a ROM, or the like. Here, the nonvolatile memory 44 is an example of a storage circuit.

Communication I/F 45 controls communication between console device 40 and other devices or systems. For example, the communication I/F 45 includes a network card, a wireless module, and the like.

The input I/F 46 receives various input operations from the examiner who operates the X-ray CT apparatus 1 . For example, the input I/F 46 includes a keyboard, a mouse, a touch panel, switches, buttons, and the like. The examiner can operate not only the console device 40 but also the entire X-ray CT apparatus 1 including the gantry device 10 and the bed device 30 using the input I/F 46. Note that the input I/F 46 is not limited to the console device 40 but may be included in the gantry device 10. For example, the inspector may manually control and move the positions of the base 31 and the top plate 32 using buttons on the gantry device 10.

The display 47 includes a display device such as an LCD (Liquid Crystal Display), and displays data such as a menu screen, a graphical user interface (GUI), a projected image, and an optical image on the display device. Note that the display 47 is not limited to the console device 40 but may be included in the gantry device 10. For example, the inspector can know the current positions of the base 31 and the top plate 32 from the display of the gantry device 10.

The reconstruction unit 51 reconstructs a tomographic image based on the projection data transmitted from the data acquisition unit 14. Specifically, the reconstruction unit 51 reconstructs the tomographic image by performing reconstruction processing on the projection data using a filtered back projection method, a successive approximation reconstruction method, or the like.

The scan control section 50 executes a scan of the subject 60 by controlling the rotation drive section 16 and the bed drive section 33 via the gantry/bed control section 17 .

With such a configuration, in the present embodiment, the scan control unit 50 performs the image acquisition unit 52, the position acquisition unit 53, the position estimation unit 54, and the movement amount calculation unit 55 before scanning the subject 60. The positioning of the subject 60 is performed by controlling.

Here, in the present embodiment, the scan control unit 50 aligns the subject 60 by aligning a position related to the examination region of the subject 60 to a target position according to the imaging conditions. In addition, in this embodiment, an example will be described in which the target position for aligning the position related to the inspection site is the rotation center of the rotating part 15.

Furthermore, in this embodiment, a coordinate system indicating a physical position in the X-ray CT apparatus 1 is referred to as a physical coordinate system, and a coordinate system indicating a position within an optical image captured by the camera 20 and the camera 21 is referred to as an image. It is called a coordinate system.

Specifically, first, the scan control unit 50 acquires an optical image of the subject 60 by controlling the camera 20 and the camera 21 via the image acquisition unit 52 and the camera control unit 22. Next, the scan control unit 50 acquires position information corresponding to the positions of a plurality of different parts of the subject 60 by inputting the optical image acquired by the image acquisition unit 52 to the position acquisition unit 53. Next, the scan control unit 50 estimates the position of the subject 60 on the bed device 30 on which the subject 60 is placed by inputting the position information acquired by the position acquisition unit 53 into the position estimation unit 54. . Furthermore, the scan control unit 50 calculates the amount of movement of the base 31 and top plate 32 of the bed device 30 by inputting the position of the subject 60 estimated by the position estimation unit 54 to the movement amount calculation unit 55. . Then, the scan control unit 50 aligns the subject 60 by moving the base 31 and the top plate 32 based on the movement amount calculated by the movement amount calculation unit 55.

The processing functions of the scan control section 50, image acquisition section 52, position acquisition section 53, position estimation section 54, and movement amount calculation section 55 will be described in detail below.

The image acquisition unit 52 acquires a first image of the subject 60 taken from a first direction, and a second image taken of the subject 60 from a second direction that is different from the first direction. do.

Specifically, the image acquisition unit 52 acquires from the camera 20, as a first image, an optical image of the subject 60 taken from diagonally above, and as a second image, An optical image of the subject 60 taken from the lateral direction is obtained from the camera 21.

The position acquisition unit 53 acquires first position information corresponding to the positions of a plurality of different parts of the subject 60 based on the first image acquired by the image acquisition unit 52, and the first position information is acquired by the image acquisition unit 52. Based on the second image, second position information corresponding to the positions of the plurality of different parts is acquired.

In the present embodiment, the position acquisition unit 53 includes, as the first position information and the second position information, a plurality of characteristic part points including the joints of the subject 60 and the position coordinates of each of the characteristic parts. Obtain skeletal information. Here, the plurality of characteristic parts including joints include, for example, the nose, neck, right shoulder, right elbow, right wrist, left shoulder, left elbow, left wrist, central buttock, right buttock, right knee, right ankle, These include the left buttock, left knee, left ankle, right eye, left eye, right ear, left ear, left thumb, left little finger, left heel, right thumb, right little finger, right heel, etc.

Here, the position acquisition unit 53 uses a learning model learned in advance using a plurality of images including the subject as learning data, and inputs the first image to the learning model to obtain the first position information. get. Further, the position acquisition unit 53 acquires second position information by inputting the second image to the learning model.

At this time, the position acquisition unit 53 uses, for example, a learning model (learning device) learned using a method of machine learning (deep learning, etc.). For example, a learning model trained in advance with a data set of a plurality of learning images including a human body as a subject and correct information of skeletal information (probability distribution of joints, etc.) in each learning image is used. As such a learning model, for example, Openpose (registered trademark) of Carnegie Mellon University is known.

Then, the position acquisition unit 53 stores the first position information and second position information (skeletal information) acquired using the learning model in the memory 43 in a format such as an array or a list. For example, with respect to a plurality of parts such as the above-mentioned joints, information indicating the distribution of the probability that each part exists in the image is output from the learning model as position information and is held in the memory 43.

In this case, for example, if the reliability distribution (probability distribution) that part n (n is an integer) exists in the image is Rn(x,y), the output R as skeleton information is R={Rn (x, y) | n=1, 2, . . . , N, N is an integer}. Here, Rn(x,y) does not have to be the reliability distribution of all regions in the image, but may be the reliability distribution of only the regions having reliability greater than a threshold value. Further, as Rn(x, y), only the peak value of reliability may be stored in association with its position coordinates (for example, part 3: right shoulder, reliability: 0.5, position coordinates: (x, y) = (122, 76) etc.). For example, in the output R, the position of the peak reliability of each part (i.e., the position where each part was detected with the highest probability of existence) was extracted, and the skeletal information was visualized based on the extracted position. An example of this is an image as shown in FIG. 6(c).

In the present embodiment, the position acquisition unit 53 performs preprocessing on images input to the learning model as preprocessing for acquiring first position information and second position information (skeletal information) using the learning model. On the other hand, image quality correction such as noise removal, distortion correction, color conversion, brightness, and color gradation correction, and correction processing such as image rotation and flipping are performed. Here, the correction parameters for performing the correction process are tabled and stored in the nonvolatile memory 44 according to the camera 20 that captures the optical image, the model of the camera 20, the imaging conditions at the time of imaging, etc. shall be. By performing correction processing using the correction parameters, the images input to the learning model are brought closer to the images of the dataset used for learning, thereby estimating skeletal information by the learning model (hereinafter referred to as skeletal estimation). can be done more accurately.

For example, an image captured in a dark room may have a tendency different from the data set used for learning due to high-sensitivity noise occurring when the image is corrected to brighten it. In such a case, it is preferable to apply processing to remove high-sensitivity noise to the input image. Similarly, if the camera 20 and the lens of the camera 20 have a wide angle and the peripheral distortion is large, it is preferable to perform distortion correction on the input image. In addition, if there are many postures of the subject in the images prepared as a data set, with the head at the top of the image and the feet at the bottom of the image, the image may be rotated or flipped to match the posture. It is preferable to input the information into the learning model. Furthermore, when learning is performed using an image that has been converted by some kind of processing, it is preferable that the input image is also similarly converted before being input to the learning model.

In addition, the position acquisition unit 53 performs projective transformation on the first image and the second image, including in the preprocessing described above, and performs projective transformation on the first image and the second image after the projective transformation. , obtain first location information and second location information.

The position estimation unit 54 estimates the position of the subject 60 on the bed device 30 on which the subject 60 is placed, based on the first position information and second position information acquired by the position acquisition unit 53.

Specifically, in the present embodiment, the position estimating unit 54 estimates the position related to the examination site of the subject 60 on the bed apparatus 30 based on the first position information and the second position information.

Here, in order to align the subject 60 by moving the base 31 and top plate 32 of the bed device 30, it is necessary to move the first It is necessary to calculate the amount of movement of the base 31 and the top plate 32 in the physical coordinate system with respect to the information and the second information (skeletal information). Therefore, the position estimating unit 54 uses the pixel resolution stored in the non-volatile memory 44 in advance to determine the image coordinates included in the first information and second information (skeletal information) acquired by the position acquiring unit 53. The position of the subject 60 on the bed apparatus 30 is estimated by converting the position coordinates of the system into the position coordinates of the physical coordinate system.

The pixel resolution used here can be determined, for example, by the following procedure. First, an object whose physical size is known in advance, such as a plate-shaped calibration plate generally used for camera calibration, is placed on the top plate 32 at a predetermined position that is designed in advance to be parallel to the top plate 32. The camera 20 takes an image of the camera 20 . Next, based on the size of the calibration plate in the captured image and the size of the physical calibration plate, the correspondence relationship between the position coordinates in the image coordinate system and the position coordinates in the physical coordinate system is determined, and the determined correspondence is determined. The relationship is stored in the nonvolatile memory 44 as the first pixel resolution. Similarly, the calibration plate is placed on the top plate 32 at a predetermined position designed in advance to be perpendicular to the top plate 32, and an image is taken by the camera 21, and the size of the calibration plate in the taken image is Based on the size of the physical calibration plate, the correspondence between the position coordinates of the image coordinate system and the position coordinates of the physical coordinate system is determined, and the determined correspondence is stored in the nonvolatile memory 44 as the second pixel resolution. I'll keep it. Note that the first pixel resolution may be determined from the known spatial arrangement between the camera 20 and each device and the specifications of the camera 20. Similarly, the second pixel resolution may be determined from the known spatial arrangement between the camera 21 and each device and the specifications of the camera 21.

The movement amount calculation unit 55 calculates the movement amount of the base 31 and the top plate 32 of the bed device 30 based on the position of the subject 60 estimated by the position estimation unit 54. Here, the movement amount calculated by the movement amount calculation section 55 is used by the scan control section 50 to move the base 31 and the top plate 32 via the gantry/bed control section 17 to align the subject 60. used for

Specifically, the movement amount calculation unit 55 calculates the movement amount of the base 31 and the top plate 32 in order to align the position related to the examination region estimated by the position estimation unit 54 to the target position according to the imaging conditions. Calculate.

The flow of positioning processing for the subject 60 performed by the X-ray CT apparatus 1 according to the first embodiment will be described below.

FIG. 2 is a flowchart showing the flow of positioning processing for the subject 60 performed by the X-ray CT apparatus 1 according to the first embodiment.

First, in step S100, the scan control unit 50 acquires preset examination plan information and subject information as imaging conditions. Here, the examination plan information includes, for example, the body orientation of the subject 60 (head first, foot first), the body position of the subject 60 (supine position, prone position, lateral position, etc.), and the subject 60's body position (supine position, prone position, lateral position, etc.). Examination site (head, chest, abdomen, etc.). Further, the subject information includes, for example, subject ID, name, sex, date of birth, age, height, weight, whether the patient is hospitalized or outpatient, and the like. This information can be acquired from RIS (Radiology Information System), HIS (Hospital Information System), etc. via communication I/F 45.

Next, in step S101, the scan control unit 50 receives setting of imaging conditions for the cameras 20 and 21 by operating the input I/F 46 by the inspector. Here, the inspector may change the imaging conditions of each camera as appropriate while displaying the images obtained from the cameras 20 and 21 on the display 47.

Next, in step S102, the scan control unit 50 performs initialization processing for the bed device 30. Here, the initialization process includes moving the base 31 and the top plate 32 so that the subject 60 can be imaged by the cameras 20 and 21 at a predesigned position. Hereinafter, the positions of the base 31 and the top plate 32 to which they are moved will be referred to as initial positions. In this initialization process, for example, the rotation center of the rotating unit 15 is set as the origin (0, 0, 0), and the reference position in the bed device 30 is spatially three-dimensional, which is linked to the image coordinate system of the camera 20 and the camera 21. The position coordinates (X0, Y0, Z0) of are stored in the nonvolatile memory 44. As a result, in subsequent steps, the position (X, Y, Z) related to the inspection site of the subject 60 with respect to the reference position is determined based on the first image and the second image obtained by the camera 20 and the camera 21. If can be estimated, the amount of movement of the rotating part 15 toward the rotation center will be (-(X0+X), -(Y0+Y), -(Z0+Z)), and by moving the base 31 and the top plate 32 by the amount of movement, The subject 60 can now be aligned to the target position.

Next, in step S103, the scan control unit 50 estimates the position of the subject 60 on the bed apparatus 30 based on the first image obtained by the camera 20 and the second image obtained by the camera 21, and estimates the position of the subject 60 on the bed apparatus 30. The automatic positioning process is executed by calculating the amount of movement of the base 31 and top plate 32 of the bed device 30 based on the position of the subject 60 thus determined. Details of the process will be described later using FIG. 3.

Next, in step S104, the scan control unit 50 may accept fine adjustment of the position of the subject 60 by manual operation of the input I/F 46 by the examiner with respect to the alignment performed in step S103. . Then, when the examiner determines that accurate alignment has been performed, the scan control unit 50 ends the alignment process. This concludes the explanation of the flowchart in FIG. 2.

FIG. 3 is a flowchart showing the flow of automatic alignment processing in step S103 shown in FIG.

In the automatic alignment process described here, the image acquisition section 52, the position acquisition section 53, and the movement amount calculation section 55 each implement the processing functions of each section in response to instructions from the scan control section 50.

First, in step S200, the image acquisition section 52 acquires a first image from the camera 20 via the camera control section 22.

FIG. 4 is a diagram illustrating an example of a first image acquired by the image acquisition unit 52 according to the first embodiment.

In this embodiment, the camera 20 is installed above the gantry device 10 so that at least a portion of the subject 60 enters the image from diagonally above. As a result, for example, as shown in FIG. 4, an optical image of the subject 60 placed on the top plate 32 taken from diagonally above is acquired as the first image.

Next, in step S201, the position acquisition unit 53 acquires first position information based on the first image acquired in step S200.

FIG. 5 is a flowchart showing the flow of the first location information acquisition process in step S201 shown in FIG. Further, FIG. 6 is a diagram showing an example of acquisition of first position information performed by the position acquisition unit 53 according to the first embodiment.

First, in step S2011, the position acquisition unit 53 performs distortion correction on the first image. Here, as described above, the position acquisition unit 53 performs noise removal, color conversion, brightness, color tone correction, etc. on the first image so that it approaches the image of the dataset used for learning. Other preprocessing such as image quality correction, image rotation, and flipping may also be performed.

Next, in step S2012, the position acquisition unit 53 performs projective transformation on the first image after distortion correction.

(a) of FIG. 6 shows an example of the first image after distortion correction, and (b) of FIG. 6 shows an example of the first image after projective transformation. As shown in FIG. 6B, projective transformation is performed on the first image after distortion correction, so that the first image after distortion correction appears as if the subject 60 was captured from directly above. image.

Finally, in step S2013, the position acquisition unit 53 uses the learning model to perform skeleton estimation on the first image after projective transformation.

(c) of FIG. 6 shows an image in which the position of the peak value of reliability of each part is plotted as the first position information obtained by skeleton estimation.

Returning to the flowchart of FIG. 3, next, in step S202, the image acquisition unit 52 acquires a second image from the camera 21 via the camera control unit 22.

FIG. 7 is a diagram illustrating an example of the second image acquired by the image acquisition unit 52 according to the first embodiment.

In this embodiment, the camera 21 is installed on the side of the gantry device 10 so that a part of the subject 60 enters the image from the side. As a result, for example, as shown in FIG. 7, an optical image captured from the lateral direction of the subject 60 placed on the top plate 32 is acquired as the second image. Here, as shown in FIG. 7, the whole body of the subject 60 does not necessarily need to be included in the second image. Furthermore, in the second image, even if the examined region is the head, the head does not necessarily need to be included in the image.

Next, in step S203, the position acquisition unit 53 acquires second position information based on the second image acquired in step S202.

FIG. 8 is a flowchart showing the flow of the second location information acquisition process in step 203 shown in FIG. Further, FIG. 9 is a diagram illustrating an example of acquisition of second position information performed by the position acquisition unit 53 according to the first embodiment.

Here, the flow of the second location information acquisition process is basically the same as the first location information acquisition process in step S201.

First, in step S2031, the position acquisition unit 53 performs distortion correction on the second image. Here, as described above, the position acquisition unit 53 applies noise removal, color conversion, brightness, color tone correction, etc. to the second image so that it approaches the image of the dataset used for learning. Other preprocessing such as image quality correction, image rotation, and flipping may also be performed.

Next, in step S2032, the position acquisition unit 53 performs projective transformation on the second image after distortion correction.

(a) of FIG. 9 shows an example of the second image after distortion correction, and (b) of FIG. 9 shows an example of the second image after projective transformation. As shown in FIG. 9B, projective transformation is performed on the second image after distortion correction, so that the second image after distortion correction appears as if the subject 60 had been imaged from the side. image.

Finally, in step S2033, the position acquisition unit 53 uses the learning model to perform skeleton estimation on the second image after projective transformation.

(c) of FIG. 9 shows an image in which the position of the peak value of reliability of each part is plotted as second position information obtained by skeleton estimation.

Returning to the flowchart of FIG. 3, next, in step S204, the position estimating unit 54 determines whether the bed device 30 is The positions of the subject 60 in the X-axis direction, Y-axis direction, and Z-axis direction are estimated.

Here, the position estimating unit 54 uses the first pixel resolution and the second pixel resolution stored in the non-volatile memory 44 in advance to determine the information contained in the first information and the second information (skeletal information). By converting the position coordinates of the image coordinate system to the position coordinates of the physical coordinate system, the positions of the subject 60 in the X-axis direction, Y-axis direction, and Z-axis direction on the bed apparatus 30 are estimated.

At this time, the position estimating unit 54 obtains the first pixel resolution and the second pixel resolution by projective transformation being performed on the first image and the second image in steps S2012 and S2032 described above. It is possible to uniformly transform the positional coordinates of the image coordinate system into the positional coordinates of the physical coordinate system for the entire image by using

Specifically, the position estimating unit 54 calculates the X-axis position information in the first position information and the Y-axis position information in the second position information with respect to the X-axis direction, Y-axis direction, and Z-axis direction that are orthogonal to each other. The position of the subject 60 in the Y-axis direction is estimated based on the position information in the direction, and the position information in the X-axis direction and the Z-axis direction in the first position information and the position information in the Y-axis direction in the second position information are estimated. Based on the position information, the position of the subject 60 on the bed apparatus 30 in the X-axis direction and the Z-axis direction is estimated.

Hereinafter, the position estimating unit 54 will be explained in more detail, taking as an example the case where the examined region is the head. When the test site is the head, the first position information and the second position information basically include the nose, right eye, left eye, and right eye as positions related to the head. It is conceivable to use positional information of parts such as the ear and the left ear.

FIG. 10 is a diagram illustrating an example of estimating the position of the subject 60 in the Y-axis direction performed by the position estimation unit 54 according to the first embodiment. Further, FIG. 11 is a diagram illustrating an example of estimating the position of the subject 60 in the X-axis direction and the Z-axis direction, which is performed by the position estimation unit 54 according to the first embodiment.

First, with respect to the Y-axis direction, it is assumed that the position related to the test region to be aligned with the target position is the position of the right ear in the second position information having the position information of the subject 60 in the Y-axis direction. In this case, in order to convert the position coordinate of the right ear in the Y-axis direction in the second position information from the image coordinate system to the physical coordinate system, the second pixel resolution stored in advance in the non-volatile memory 44 is used. There is a need.

However, as shown in FIG. 10, for example, if the second pixel resolution is acquired on the plane P0, the pixel resolution on the plane P1 passing through the right ear position is the same as that between the camera 21 and the plane P0. Assuming that the distance between the planes P0 and P1 is L0, and the distance between the planes P0 and P1 is L1, it is appropriate to set it to (L0-L1)/L0*(second pixel resolution). Here, L0 and L1 can be calculated from the position of the right ear in the X-axis direction according to the first position information, as shown in FIG. 10 in which the first position information is superimposed on the schematic diagram.

Therefore, the position estimation unit 54 converts the position information in the Y-axis direction in the second position information using the pixel resolution based on the position information in the X-axis direction in the first position information. The position of the subject 60 in the Y-axis direction is estimated.

Specifically, as described above, the position estimating unit 54 calculates L0 and L1 from the position of the right ear in the X-axis direction in the first position information, and uses the L0 and L1 to perform the acquisition on the plane P0. The pixel resolution on the plane P1 is calculated from the second resolution obtained. Then, the position estimating unit 54 uses the calculated pixel resolution in the plane P1 to convert the position coordinate of the right ear in the Y-axis direction in the second position information from the image coordinate system to the physical coordinate system. The position of the right ear of the subject 60 in the device 30 in the Y-axis direction is estimated.

In this way, by estimating the position of the subject 60 in the Y-axis direction using both the first position information and the second position information, the position of the subject 60 can be more accurately aligned in the Y-axis direction. You will be able to do this.

Although L1 is calculated here using the position of the right ear in the first position information, the method for calculating L1 is not limited to this. For example, a distance twice the distance between the position of the right eye in the X-axis direction and the plane P0 in the first position information may be calculated as L1. That is, L1 may be a distance calculated based on a position related to the inspection site in the first position information.

In addition, here, we used the positional information of parts such as the nose, right eye, left eye, right ear, and left ear as positions related to the head, but the positional information used when the test part is the head is , is not necessarily limited to positional information of parts related to the head. For example, position information obtained by estimating the position of the head from position information of the neck and waist may be used.

Next, with respect to the X-axis direction and the Z-axis direction, the position related to the test region to be aligned to the target position is the nose in the first position information having the position information of the subject 60 in the X-axis direction and the Z-axis direction; It is assumed that this is the center of gravity of the positional coordinates of parts such as the right eye, left eye, right ear, and left ear.

Here, the position coordinates of each part in the first position information are determined by projective transformation on the first image in step S2012 described above, so that the position coordinates of each part in the first position information are determined on a single plane, that is, at the first pixel resolution. It is calculated from the image projected onto the acquired plane. Therefore, the positional coordinates of each part in the first positional information include distortion according to the height in the Y-axis direction from the top plate 32, which is the projection plane.

For example, as shown in FIG. 11, when the subject 60 is viewed from the side, the position of the camera 20 is (0, H0), the position of the center of gravity is (Z, H1), and the pixel at the center of gravity is If the position projected onto the projection plane P2 is (Z+Ez, 0), the error Ez from the position Z to be determined is Ez=Z*H1/(H0-H1). This can be considered similarly for the X-axis direction, so the error Ex from the position X to be determined is Ex=X*H1/(H0-H1). Here, H0 and H1 can be calculated from the position of the center of gravity in the Y-axis direction in the second position information.

Therefore, the position estimation unit 54 calculates the height from the top plate 32 based on the position information in the X-axis direction and the Z-axis direction in the first position information and the position information in the Y-axis direction in the second position information. Then, the position of the subject 60 in the X-axis direction and the Z-axis direction on the bed apparatus 30 is estimated.

Specifically, as described above, the position estimation unit 54 calculates H0 and H1 from the position of the center of gravity in the Y-axis direction in the second position information, and uses the H0 and H1 to calculate the position of the center of gravity in the Z-axis direction. An error Ez and an error Ex in the X-axis direction are calculated. In addition, the position estimating unit 54 subtracts the error Ez and the error Ex from the position coordinates of the center of gravity in the Z-axis direction and the X-axis direction in the first position information, respectively, thereby calculating the Z-axis direction and the Calculate the position coordinates of the direction. Then, the position estimation unit 54 converts the calculated position coordinates in the Z-axis direction and the X-axis direction from the image coordinate system to the physical coordinate system using the first pixel resolution, thereby determining the center of gravity of the bed device 30. Estimate the position in the X-axis direction.

In this way, by estimating the position in the X-axis direction and the Z-axis direction using both the first position information and the second position information, the position of the subject 60 can be adjusted in the X-axis direction and the Z-axis direction. can be done more accurately.

In addition, here, we used the center of gravity of the positional coordinates of the nose, right eye, left eye, right ear, and left ear as positions related to the head, but the center of gravity used here does not necessarily include all of these points. It does not have to be calculated using the position coordinates of the part. For example, the center of gravity may be calculated after selecting parts based on the peak value of reliability. Alternatively, the center of gravity may be calculated by multiplying the position coordinates by a coefficient based on the peak value of reliability. Alternatively, the center of gravity may be calculated after weighting the position coordinates for each part.

Returning to the flowchart in FIG. 3, in step S205, the movement amount calculation unit 55 moves the couch device based on the positions of the subject 60 in the X-axis direction, Y-axis direction, and Z-axis direction estimated in step S204. The amount of movement of the base 31 and top plate 32 of No. 30 in the X-axis direction, Y-axis direction, and Z-axis direction is calculated.

For example, in the example shown in FIG. 10, the movement amount calculation unit 55 moves the right ear of the subject 60 to the rotation unit 15, which is the target position, based on the position of the right ear in the Y-axis direction estimated by the position estimation unit 54. The amount of movement of the base 31 in the Y-axis direction for alignment with the center of rotation is calculated.

For example, in the example shown in FIG. 11, the movement amount calculation unit 55 targets the right ear of the subject 60 based on the position of the center of gravity in the X-axis direction and the Z-axis direction estimated by the position estimation unit 54. The amount of movement of the top plate 32 in the X-axis direction and the Z-axis direction for alignment with the rotation center of the rotating unit 15 is calculated.

Next, in step S206, the scan control unit 50 controls the gantry/bed control unit 17 to move the bed device 30 based on the movement amounts in the X-axis direction, Y-axis direction, and Z-axis direction calculated in step S205. By moving the base 31 and top plate 32, the subject 60 is aligned in the X-axis direction, Y-axis direction, and Z-axis direction.

Note that in the flowchart of FIG. 3, after acquiring the first image and first position information in steps S200 to S201, the second image and second position information are acquired in steps S202 to S203. However, the embodiment is not limited to this. For example, the processing in steps S200 to S201 and the processing in steps S202 to S203 may be performed in a reverse order, or may be performed in parallel.

The processing functions of the scan control section 50, image acquisition section 52, position acquisition section 53, position estimation section 54, and movement amount calculation section 55 have been described above. It is executed in each section according to the execution or instructions from the processing circuit 42.

Here, when the processing functions of each part are executed by the processing circuit 42, for example, the processing circuit 24 is realized by a processor. Further, when the processing functions of each section are executed by each section according to instructions from the processing circuit 42, for example, the scan control section 50, the image acquisition section 52, the position acquisition section 53, the position estimation section 54, and the movement amount calculation section 55 are each Realized by a processor. In these cases, the processing functions of the scan control section 50, image acquisition section 52, position acquisition section 53, position estimation section 54, and movement amount calculation section 55 are stored in the nonvolatile memory 44 in the form of a computer-executable program. be remembered. Then, the processing circuit 24 or the scan control section 50, image acquisition section 52, position acquisition section 53, position estimation section 54, and movement amount calculation section 55 read each program from the nonvolatile memory 44 and execute it. Realize processing functions corresponding to each program. In other words, the console device 40 has the configuration shown in FIG. 1 in a state where the processing circuit 24 or each section has read each program.

As described above, according to the first embodiment, the first position information acquired based on the first image taken from the first direction, and the second position information that is in a direction different from the first direction. By estimating the position of the subject 60 on the couch device 30 using both the second position information acquired based on the second image taken from the direction, the position of the subject can be more accurately aligned. It can be carried out.

(Second embodiment)
Note that in the first embodiment described above, an example has been described in which alignment of the subject 60 in the X-axis direction, Y-axis direction, and Z-axis direction is performed simultaneously, but the embodiment is not limited to this. For example, positioning of the subject 60 in the X-axis direction and Z-axis direction and positioning of the subject 60 in the Y-axis direction may be performed separately. An example of such a case will be described below as a second embodiment. Note that, in the following description, points that are different from the first embodiment will be mainly explained, and detailed descriptions of contents common to the first embodiment will be omitted.

FIG. 12 is a flowchart showing the flow of automatic alignment processing according to the second embodiment.

Here, the flowchart shown in FIG. 12 corresponds to the automatic positioning process of step S103 shown in FIG. 2 in the first embodiment, and is a different embodiment from the flowchart shown in FIG. 3.

First, in step S300, the image acquisition unit 52 acquires a first image from the camera 20 via the camera control unit 22, similar to the process in step S200 shown in FIG.

Next, in step S301, the position acquisition unit 53 acquires first position information based on the first image acquired in step S300, similar to the process in step S201 shown in FIG.

Next, in this embodiment, in step S302, the position estimating unit 54 estimates the position of the subject 60 in the X-axis direction and the Z-axis direction on the bed apparatus 30 based on the first position information acquired in step S301. Estimate.

Here, the position estimation unit 54 uses the first pixel resolution stored in the nonvolatile memory 44 in advance to calculate the first information and the second information (skeletal information) acquired by the position acquisition unit 53. By converting the position coordinates of the included image coordinate system into the position coordinates of the physical coordinate system, the position of the subject 60 in the X-axis direction and the Z-axis direction on the bed device 30 is estimated.

At this time, as explained in the first embodiment, the position coordinates of each part in the first position information include distortion according to the height in the Y-axis direction from the top plate 32, which is the projection plane. However, here, the position of the subject 60 in the X-axis direction and the Z-axis direction is estimated with distortion included. In this embodiment, in step S308, which will be described later, the positions of the subject 60 in the X-axis direction and the Z-axis direction are corrected by estimating the positions in the X-axis direction and the Z-axis direction again.

Next, in step S303, the movement amount calculation unit 55 calculates the The amount of movement in the axial direction, Y-axis direction, and Z-axis direction is calculated.

Next, in step S304, the scan control unit 50 controls the gantry/bed control unit 17 to move the top plate 32 based on the amount of movement in the X-axis direction and the Z-axis direction calculated in step S303. Accordingly, the subject 60 is aligned in the X-axis direction and the Z-axis direction.

Next, in step S305, the image acquisition unit 52 acquires a second image from the camera 21 via the camera control unit 22, similar to the process in step S202 shown in FIG.

Next, in step S306, the position acquisition unit 53 acquires second position information based on the second image acquired in step S305, similar to the process in step S203 shown in FIG.

Next, in step S307, the position estimating unit 54 determines the X-axis of the subject 60 on the couch device 30 based on the first position information acquired in step S301 and the second position information acquired in step S307. The direction, the position in the Y-axis direction, and the Z-axis direction are estimated.

Here, the process in step S307 is almost the same as the process in step S204 shown in FIG. 3, but in this embodiment, in step S304, based on the movement amount calculated in step S303, The top plate 32 is also moved in the Z-axis direction. Therefore, here, the position estimating unit 54 estimates only the errors Ex and Ez due to the distortion according to the height in the Y-axis direction from the top plate 32 as described above for the positions in the X-axis direction and the Z-axis direction. do it. Note that the position estimating unit 54 estimates the position in the Y-axis direction in the same manner as in the process of step S204 shown in FIG. 3.

Next, in step S308, the movement amount calculation unit 55 moves the base 31 and the top of the bed device 30 based on the positions of the subject 60 in the X-axis direction, Y-axis direction, and Z-axis direction estimated in step S307. The amount of movement of the plate 32 in the X-axis direction, Y-axis direction, and Z-axis direction is calculated.

Next, in step S309, similarly to step S206 shown in FIG. 3, the scan control unit 50 moves the gantry and By controlling the bed control unit 17 to move the base 31 and top plate 32 of the bed device 30, the subject 60 is aligned in the X-axis direction, Y-axis direction, and Z-axis direction.

As described above, according to the second embodiment, by separately performing the alignment of the subject 60 in the X-axis direction and the Z-axis direction and the alignment of the subject 60 in the Y-axis direction, for example, Even when the space that can be imaged by the cameras 20 and 21 is not common, positioning of the subject 60 can be performed.

(Modified example)
In the above-described embodiment, as shown in FIG. 5, the position acquisition unit 53 performs projective transformation on the image after distortion correction, and then uses the learning model to However, the embodiment is not limited to this. For example, when the resolution of the cameras 20 and 21 is high, or when the accuracy of the learning model is high, when it is possible to perform skeleton estimation even from images before projective transformation, the position acquisition unit 53 may be configured to perform skeletal estimation on the distortion-corrected image using the learning model, and then perform projective transformation on the distortion-corrected image.

Further, in the embodiment described above, the position estimating unit 54 determines whether or not the bed Although the position related to the test site of the subject 60 in the apparatus 30 is estimated, the position related to the test site does not necessarily have to be included in both the first image and the second image. . For example, if one of the first image and the second image does not include a position related to the examination site, the position estimation unit 54 calculates the position obtained based on the other image. The position related to the test site may be estimated based on the position information of other sites in the information.

In addition, in the embodiment described above, an example has been described in which the target position for aligning the position related to the examination region in order to align the subject 60 is the rotation center of the rotating unit 15. It is not limited to this.

For example, the scan performed after positioning of the subject 60 is performed using the method of the embodiment described above may be a preparatory scan for capturing a 2D or 3D scanogram before the main scan. , or may be a main scan for photographing images for diagnosis. Here, regardless of which scan is performed, the target position for aligning the position related to the inspection site is determined according to the imaging conditions used in the scan.

For example, the bed device 3 In the case of scanning performed with the top plate 32 fixed in position, the target position is the center of rotation of the rotating unit 15, as in the embodiment described above.

Furthermore, in the case of a scan performed while moving the top plate 32 of the bed device 30, such as a 2D helical scan or a 3D helical scan, the top plate 32 is usually accelerated to move at a constant speed. Since scanning is started after the top plate 32 starts moving, a run-up section is required from when the top plate 32 starts moving until it moves at a constant velocity. Therefore, when such a scan is performed, the target position is set, for example, in consideration of the run-up section, and is set at a position that is the distance of the run-up section from the end of the photographing range in the gantry device 10. Note that by performing a helical scan (3D landmark scan) without stopping the bed device 30 while moving the base 31 and the top plate 32 for positioning the subject 60 in the method of the above-described embodiment, If the run-up section is not required, the target position may be the center of rotation of the rotating section 15. Also, for example, when estimating the positions of multiple examination parts (e.g., head and abdomen) and scanning each examination part sequentially, for the first examination part (e.g., head), The target position is set in consideration of the run-up section, and for the second inspection area (for example, the abdomen), the top plate 32 has already moved at a constant speed to scan the first inspection area, and the run-up section is set as the target position. Since this is not necessary, the target position is the center of rotation of the rotating section 15.

Further, in the embodiment described above, the scan control unit 50 controls the substrate 31 and the top plate 32 of the bed device 30 based on the first image captured by the camera 20 and the second image captured by the camera 21. Although an example of moving has been described, the embodiment is not limited to this. For example, if the X-ray CT apparatus 1 is configured to perform imaging by tilting the gantry device 10, the scan control unit 50 controls the first image captured by the camera 20 and the camera The gantry device 10 may be tilted based on the second image captured by the gantry 21.

For example, when the examination site is the head, the gantry device 10 is tilted so that the photographed cross section passes through the OM (OrbitoMetal) line, which is the line connecting the ear canal and the external canthus. There is something to do. In such a case, for example, the position estimating unit 54 determines the position of the OM line of the subject 60 on the bed apparatus 30 based on the first position information and the second position information acquired by the position acquiring unit 53. presume. In this case, the position of the OM line is an example of a position related to the examination site. Furthermore, based on the position of the OM line estimated by the position estimating unit 54, the movement amount calculating unit 55 calculates an inclination angle for inclining the gantry device 10 so that the photographed cross section passes through the OM line. In this case, the position of the photographed cross section is an example of a target position. Then, the scan control unit 50 tilts the gantry device 10 by the tilt angle calculated by the movement amount calculation unit 55.

Generally, when such imaging is performed, a scano image is taken before the main scan, and the gantry device 10 is tilted based on the OM line determined using the scano image. be exposed. On the other hand, according to the above configuration, the gantry device 10 can be tilted based on the first image captured by the camera 20 and the second image captured by the camera 21, and the scano image It is now possible to take pictures with the gantry device 10 tilted without having to take pictures.

Further, in the embodiment described above, the movement amount calculation unit 55 calculates the movement amount of the base 31 and the top plate 32 of the bed device 30 based on the position of the subject 60 estimated by the position estimation unit 54, Although the scan control unit 50 moves the base 31 and the top plate 32 based on the movement amount calculated by the movement amount calculation unit 55, the embodiment is not limited to this. For example, the movement amount calculation unit 55 is configured to determine the position to which the base 31 and the top plate 32 of the bed apparatus 30 are to be moved based on the position of the subject 60 estimated by the position estimation unit 54, and The control unit 50 may be configured to move the base 31 and the top plate 32 to the positions determined by the movement amount calculation unit 55.

In addition, in the embodiment described above, an example is shown in which the subject 60 is placed head-first in the bed device 30 in which the head is placed on the side of the gantry device 10 in FIG. It is not limited to this. For example, the embodiment described above can be similarly applied even when the subject 60 is placed foot first with the legs placed on the gantry device 10 side.

(Other embodiments)
Note that in the embodiment described above, an example was described in which the processing functions of the image acquisition unit, position acquisition unit, position estimation unit, and movement amount calculation unit in this specification are realized by a program, but the embodiment is not limited to this. I can't do it. For example, the image acquisition unit, position acquisition unit, position estimation unit, and movement amount calculation unit in this specification may realize each processing function using only hardware, only software, or a mixture of hardware and software. I don't mind.

Further, in the embodiment described above, the scan control unit 50, the image acquisition unit 52, the position acquisition unit 53, the position estimation unit 54, and the movement amount calculation unit 55 are realized by a single processor, but in the embodiment is not limited to this. For example, the scan control unit 50, image acquisition unit 52, position acquisition unit 53, position estimation unit 54, and movement amount calculation unit 55 are configured by combining a plurality of independent processors, and each processor executes a program to It may also be something that realizes a processing function. In addition, each processing function of the scan control unit 50, image acquisition unit 52, position acquisition unit 53, position estimation unit 54, and movement amount calculation unit 55 is realized by being distributed or integrated into a single or multiple processing units as appropriate. may be done. Further, in the embodiment described above, programs corresponding to each processing function are stored in a single nonvolatile memory 44, but the embodiment is not limited to this. For example, a configuration may be adopted in which programs corresponding to each processing function are stored in a distributed manner in a plurality of storage circuits, and are read out from each storage circuit and executed.

Furthermore, the term "processor" used in the description of the embodiments described above refers to, for example, a CPU, a GPU (Graphics Processing Unit), an application specific integrated circuit (ASIC), a programmable logic device (for example, a simple Refers to circuits such as a programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA). When the processor is, for example, a CPU, the processor realizes processing functions by reading and executing a program stored in a storage circuit. On the other hand, when the processor is an ASIC, instead of storing the program in a storage circuit, the processing function is directly incorporated into the processor's circuitry as a logic circuit. Note that each processor in this embodiment is not limited to being configured as a single circuit for each processor, but may also be configured as a single processor by combining multiple independent circuits to realize its functions. good. Furthermore, a plurality of components shown in FIG. 1 may be integrated into one processor to realize its functions.

Here, the program executed by the processor is provided by being pre-installed in a storage circuit such as an HDD or ROM, for example. This program is a file in a format that can be installed on these storage circuits or an executable format, such as a CD (Compact Disk)-ROM, FD (Flexible Disk), CD-R (Recordable), or DVD (Digital Versatile Disk). It may be recorded and provided on a computer-readable non-transitory storage medium such as. Further, this program may be stored on a computer connected to a network such as the Internet, and may be provided or distributed by being downloaded via the network. For example, this program is composed of modules including each of the processing functions described above. In actual hardware, a CPU reads a program from a storage medium such as a ROM and executes it, so that each module is loaded onto the main storage device and generated on the main storage device.

Furthermore, in the embodiments described above, each component of each device illustrated is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distributing or integrating each device is not limited to what is shown in the diagram, and all or part of the devices may be functionally or physically distributed or integrated in arbitrary units depending on various loads and usage conditions. Can be integrated and configured. Further, each processing function performed by each device may be realized in whole or in part by a CPU and a program that is analyzed and executed by the CPU, or may be realized as hardware using wired logic.

Furthermore, among the processes described in the above-described embodiments, all or part of the processes described as being performed automatically can be performed manually, or the processes described as being performed manually can be performed manually. All or part of this can also be performed automatically using known methods. In addition, information including processing procedures, control procedures, specific names, and various data and parameters shown in the above documents and drawings may be changed arbitrarily, unless otherwise specified.

According to at least one embodiment described above, it is possible to align the subject more accurately.

Although several embodiments have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and changes can be made without departing from the gist of the invention. These embodiments and their modifications are included within the scope and gist of the invention as well as within the scope of the invention described in the claims and its equivalents.

1 X-ray CT device 52 Image acquisition unit 53 Position acquisition unit 54 Position estimation unit 55 Movement amount calculation unit

Claims (10)

an image acquisition unit that acquires a first image of the subject taken from a first direction and a second image of the subject taken from a second direction that is different from the first direction;
First position information corresponding to the positions of the plurality of different parts of the subject is acquired based on the first image, and first position information corresponding to the positions of the plurality of different parts is obtained based on the second image. a position acquisition unit that acquires second position information;
An X-ray computed tomography apparatus, comprising: a position estimation unit that estimates the position of the subject on a bed on which the subject is placed, based on the first position information and the second position information. further comprising a movement amount calculation unit that calculates a movement amount of the bed based on the position of the subject;
The X-ray computed tomography apparatus according to claim 1. The position acquisition unit obtains the first position information by inputting the first image into the learning model using a learning model trained in advance using a plurality of images including the subject as learning data. acquiring the second position information by inputting the second image into the learning model;
The X-ray computed tomography apparatus according to claim 1 or 2. The position acquisition unit obtains, as the first position information and the second position information, skeletal information including a plurality of characteristic part points including joints of the subject and position coordinates of each of the characteristic parts. get,
The X-ray computed tomography apparatus according to claim 1 or 2. The position estimation unit is configured to calculate the position information in the X-axis direction in the first position information and the Y-axis direction in the second position information with respect to the X-axis direction, Y-axis direction, and Z-axis direction that are orthogonal to each other. The position of the subject in the Y-axis direction is estimated based on the position information in the X-axis direction and the Z-axis direction in the first position information, and in the second position information. estimating the position of the subject on the bed in the X-axis direction and the Z-axis direction based on the position information in the Y-axis direction;
The X-ray computed tomography apparatus according to claim 1 or 2. The position estimating unit converts the position information in the Y-axis direction in the second position information using pixel resolution based on the position information in the X-axis direction in the first position information. estimating the position of the subject in the Y-axis direction;
The X-ray computed tomography apparatus according to claim 5. The position estimating unit calculates the height from the bed based on the position information in the X-axis direction and the Z-axis direction in the first position information, and the position information in the Y-axis direction in the second position information. estimating the position of the subject on the bed in the X-axis direction and the Z-axis direction based on
The X-ray computed tomography apparatus according to claim 5. The position acquisition unit performs projective transformation on the first image and the second image, and acquires the first position information and the second image based on the first image and second image after the projective transformation. acquiring the second location information;
The X-ray computed tomography apparatus according to claim 1 or 2. The position estimating unit estimates a position related to the test site of the subject on the bed based on the first position information and the second position information,
The movement amount calculation unit calculates a movement amount of the bed for aligning a position related to the examination region to a target position according to imaging conditions.
The X-ray computed tomography apparatus according to claim 2. An estimation method for estimating the position of a subject on a bed of an X-ray computed tomography apparatus, comprising:
acquiring a first image of the subject taken from a first direction and a second image of the subject taken from a second direction that is different from the first direction;
First position information corresponding to the positions of the plurality of different parts of the subject is acquired based on the first image, and first position information corresponding to the positions of the plurality of different parts is obtained based on the second image. obtaining second location information;
and estimating the position of the subject on the bed on which the subject is placed, based on the first position information and the second position information.

Images