Classifications

• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
  • A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
  • A61B6/025—Tomosynthesis
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
  • A61B6/50—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications
  • A61B6/502—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications for diagnosis of breast, i.e. mammography
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
  • A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
  • A61B6/5205—Devices using data or image processing specially adapted for radiation diagnosis involving processing of raw data to produce diagnostic data
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
  • A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
  • A61B6/5258—Devices using data or image processing specially adapted for radiation diagnosis involving detection or reduction of artifacts or noise
  • A61B6/5264—Devices using data or image processing specially adapted for radiation diagnosis involving detection or reduction of artifacts or noise due to motion
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T7/00—Image analysis
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T7/00—Image analysis
  • G06T7/0002—Inspection of images, e.g. flaw detection
  • G06T7/0012—Biomedical image inspection
• • G—PHYSICS
  • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
  • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
  • G16H30/00—ICT specially adapted for the handling or processing of medical images
  • G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
  • A61B6/04—Positioning of patients; Tiltable beds or the like
  • A61B6/0407—Supports, e.g. tables or beds, for the body or parts of the body
  • A61B6/0414—Supports, e.g. tables or beds, for the body or parts of the body with compression means
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/10—Image acquisition modality
  • G06T2207/10072—Tomographic images
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2211/00—Image generation
  • G06T2211/40—Computed tomography
  • G06T2211/436—Limited angle

Definitions

• tomosynthesis imaging also has a problem in that the reconstructed tomographic image becomes blurred due to the influence of body movements of the subject due to the time difference between imaging at each of a plurality of radiation source positions.
• a tomographic image is blurred in this way, it becomes difficult to discover lesions such as minute calcifications, which are useful for early detection of breast cancer, especially when the subject is a breast.
• feature points are detected from tomographic images, but feature points are not always detected on projected images. For example, even if a feature point exists with high contrast in a tomographic image, the contrast of the feature point may be low in a projection image, making it difficult to detect the feature point. If the feature points are difficult to detect in the projection image, the feature points on the tomographic image and the feature points on the projection image cannot be accurately correlated, and in this case, body movements cannot be accurately corrected.
• the present disclosure has been made in view of the above circumstances, and provides an image processing device, method, and program that can obtain high-quality tomographic images in which body movements are accurately corrected.
• An image processing device includes at least one processor, The processor is Multiple images generated by moving the radiation source relative to the detection surface of the detection unit and having the imaging device perform tomosynthesis imaging in which the subject is irradiated with radiation at multiple source positions due to the movement of the radiation source.
• Obtain multiple projection images corresponding to each source position Extract specific structures from multiple projection images and derive multiple structural projection images, Reconstruct multiple structural projection images to derive structural tomographic images for each of the multiple tomographic planes of the object, detecting at least one characteristic structure from a plurality of structural tomographic images; In the corresponding tomographic plane corresponding to the structural tomographic image in which the characteristic structure has been detected, using the characteristic structure as a reference, correcting the positional deviation between the plurality of projection images based on the body movement of the subject and reconstructing the plurality of projection images;
• the image processing apparatus is configured to derive a corrected tomographic image in at least one tomographic plane of the subject.
• “Moving the radiation source relative to the detection section” refers to moving only the radiation source, only the detection section, or moving both the radiation source and the detection section. Also included.
• the specific structure may be at least one of a line structure and a point structure.
• the processor may extract at least one of the line structure and the point structure based on the degree of concentration of gradient vectors representing the gradient of pixel values in the projected image.
• the processor derives the amount of positional deviation between the plurality of projected images based on the body movement of the subject in the corresponding tomographic plane, using the feature structure as a reference,
• a corrected tomographic image may be derived by correcting the amount of positional deviation and reconstructing a plurality of projection images.
• the processor detects a plurality of characteristic structures from a plurality of structural tomographic images, determining whether a corresponding tomographic plane corresponding to a structural tomographic image in which each of the plurality of characteristic structures is detected is a focal plane; The amount of positional deviation may be derived in the corresponding tomographic plane determined to be the focal plane.
• the processor may detect points that satisfy a specific threshold condition in the structural tomographic image as characteristic structures.
• the processor updates the structural tomographic image by reconstructing the structural projection image while correcting the positional shift; Detect the updated feature structure from the updated structural tomographic image, Update the amount of positional deviation using the updated feature structure, The update of the structural tomographic image, the feature structure, and the amount of positional deviation may be repeated.
• the processor updates the structural tomographic image by reconstructing the structural projection image while correcting the positional shift; detecting an updated feature structure from the updated structural tomographic image based on the updated threshold condition; Update the amount of positional deviation using the updated feature structure, The updating of the structural tomographic image, the updating of the feature structure based on the updated threshold condition, and the updating of the positional deviation amount may be repeated.
• the processor projects the plurality of projection images onto the corresponding tomographic plane based on the positional relationship between the radiation source position and the detection unit at the time of imaging for each of the plurality of projection images. Then, a tomographic projection image corresponding to each of the plurality of projection images is derived, In the corresponding tomographic plane, the amount of positional deviation between the plurality of tomographic projection images based on the body movement of the subject may be derived as the amount of positional deviation between the plurality of projection images using the feature structure as a reference.
• the processor may set a local region corresponding to the feature structure in the plurality of tomographic projection images, and derive the positional deviation amount based on the local region. .
• a "local region” is a region that includes a characteristic structure in a tomographic image or a tomographic projection image, and can be a region of any size that is smaller than the tomographic image or tomographic projection image.
• the local area needs to be larger than the range of body movement. If the body movement is large, it may be about 2 mm. Therefore, in the case of a tomographic image or a tomographic projection image in which each pixel has a size of 100 ⁇ m square, the local region may be, for example, a region of 50 ⁇ 50 pixels or 100 ⁇ 100 pixels around the characteristic structure.
• a region around a feature structure in a local region means a region smaller than the local region that includes the feature structure in the local region.
• the processor sets a plurality of first local regions including the characteristic structure in the plurality of tomographic projection images, and sets the plurality of first local regions including the characteristic structure in the tomographic image in which the characteristic structure is detected.
• a second local area including a plurality of first local areas is set, and the positional deviation amounts of the plurality of first local areas with respect to the second local area are respectively derived as temporary positional deviation amounts, and the position is calculated based on the plurality of temporary positional deviation amounts. It may also be a method for deriving the amount of deviation.
• the processor may derive the temporary positional shift amount based on the area around the feature structure in the second local area.
• the processor reconstructs the plurality of projection images excluding the target projection image corresponding to the target tomographic plane projection image from which the amount of positional deviation is to be derived, and generates a plurality of tomographic images. is derived as the target tomographic image,
• the positional shift amount for the target tomographic plane projection image may be derived using the target tomographic image.
• the processor evaluates the image quality of the region of interest including the feature structure in the corrected tomographic image, and determines whether the derived positional deviation amount is appropriate based on the result of the image quality evaluation. It may also be used to determine whether it is inappropriate.
• the processor derives a plurality of tomographic images by reconstructing a plurality of projection images, Evaluate the image quality of the region of interest including the characteristic structure in the tomographic image, compare the image quality evaluation results for the corrected tomographic image with the image quality evaluation results for the tomographic image, and select the tomographic image with the better image quality evaluation result. may be determined as the final tomographic image.
• the processor derives an evaluation function for evaluating the image quality of a region of interest including a feature structure in the corrected tomographic image, Alternatively, the amount of positional deviation that optimizes the evaluation function may be derived.
• the subject may be a breast.
• the processor performs processing according to at least one of breast density, breast size, imaging time of tomosynthesis imaging, breast compression pressure during tomosynthesis imaging, and breast imaging direction.
• the search range when deriving the amount of positional deviation may be changed.
• An image processing method moves a radiation source relative to a detection surface of a detection unit, and causes an imaging device to perform tomosynthesis imaging in which a subject is irradiated with radiation at a plurality of source positions by moving the radiation source.
• An image processing program causes an imaging device to perform tomosynthesis imaging in which a radiation source is moved relative to a detection surface of a detection unit and a subject is irradiated with radiation at a plurality of radiation source positions by moving the radiation source.
• a computer is caused to perform processing including deriving a corrected tomographic image in at least one tomographic plane of the subject.
• a schematic configuration diagram of a radiographic imaging apparatus to which an image processing apparatus according to a first embodiment of the present disclosure is applied Diagram of the radiographic imaging device viewed from the direction of arrow A in Figure 1
• a diagram showing a schematic configuration of an image processing device according to the first embodiment A diagram showing the functional configuration of an image processing device according to the first embodiment Diagram for explaining the acquisition of projection images Diagram showing projected images and structural projected images Diagram for explaining derivation of structural tomographic images Diagram for explaining detection of feature structures from structural tomographic images Diagram for explaining projection of a projection image onto a corresponding tomographic plane Diagram for explaining interpolation of pixel values of tomographic images Diagram to explain setting the region of interest Diagram showing the region of interest set on the tomographic projection image
• a diagram showing an image within the region of interest when no body movement occurs in the first embodiment A diagram showing an image within the region of interest when body movement occurs in the first embodiment Diagram to explain the search range of the region of interest Diagram showing feature structure in three-dimensional space Diagram showing the display screen of corrected tomographic images Flowchart showing
• FIG. 1 is a schematic configuration diagram of a radiographic imaging system to which an image processing apparatus according to a first embodiment of the present disclosure is applied.
• a radiation imaging system 1 photographs the breast, which is a subject, from a plurality of radiation source positions, and generates a plurality of radiographic images, i.e. This is for acquiring multiple projection images.
• the radiographic imaging system 1 includes an imaging device 10, a console 2, an image storage system 3, and an image processing device 4 according to this embodiment.
• FIG. 2 is a view of the imaging device in the radiographic imaging system as seen from the direction of arrow A in FIG.
• the imaging device 10 is a mammography system that photographs a breast M, which is a subject, from a plurality of radiation source positions to obtain a plurality of radiation images, that is, a plurality of projection images, in order to perform tomosynthesis imaging of the breast and derive a tomographic image. It is a photographic device.
• the photographing device 10 includes an arm portion 12 connected to a base (not shown) by a rotating shaft 11.
• An imaging table 13 is attached to one end of the arm section 12, and a radiation irradiation section 14 is attached to the other end so as to face the imaging table 13.
• the arm portion 12 is configured such that only the end portion to which the radiation irradiation unit 14 is attached can be rotated, and thereby, it is possible to fix the imaging table 13 and rotate only the radiation irradiation unit 14. It has become.
• the rotation of the arm portion 12 is controlled by the console 2.
• a radiation detector 15 such as a flat panel detector is provided inside the imaging table 13.
• the radiation detector 15 has a detection surface 15A for detecting radiation such as X-rays.
• a charge amplifier that converts the charge signal read out from the radiation detector 15 into a voltage signal
• a correlated double sampling circuit that samples the voltage signal output from the charge amplifier
• a voltage signal There are also circuit boards equipped with an AD (Analog Digital) converter, etc. that converts signals into digital signals.
• AD Analog Digital
• the radiation detector 15 is an example of a detection section. Further, in this embodiment, the radiation detector 15 is used as the detection unit, but the detection unit is not limited to the radiation detector 15 as long as it can detect radiation and convert it into an image.
• the radiation detector 15 is capable of repeatedly recording and reading radiation images, and may be a so-called direct type radiation detector that directly converts radiation such as X-rays into electric charges, or may be a A so-called indirect radiation detector may be used, which first converts the radiation into visible light and then converts the visible light into a charge signal.
• a readout method for radiation image signals there is a so-called TFT readout method in which radiation image signals are read out by turning on and off a TFT (Thin Film Transistor) switch, or a radiation image signal is read out by irradiating reading light.
• TFT Thin Film Transistor
• An X-ray source 16 which is a radiation source, is housed inside the radiation irradiation unit 14.
• the console 2 controls the timing of irradiating X-ray radiation from the X-ray source 16 and the X-ray generation conditions in the X-ray source 16, such as selection of target and filter materials, tube voltage, and irradiation time.
• the arm part 12 includes a compression plate 17 that is arranged above the imaging table 13 and presses down the breast M, a support part 18 that supports the compression plate 17, and a support part 18 that is arranged above and below in FIGS. 1 and 2.
• a moving mechanism 19 for moving in the direction is provided.
• the console 2 receives imaging orders and various information obtained from an unillustrated RIS (Radiology Information System), etc., as well as instructions directly given by technicians, etc. via a network such as a wireless communication LAN (Local Area Network). It has a function of controlling the photographing device 10 using the camera. Specifically, the console 2 causes the imaging device 10 to perform tomosynthesis imaging of the breast M, thereby acquiring a plurality of projection images as described later. As an example, in this embodiment, a server computer is used as the console 2.
• RIS Radiology Information System
• LAN Local Area Network
• the image storage system 3 is a system that stores image data such as radiation images and tomographic images taken by the imaging device 10.
• the image storage system 3 extracts image data corresponding to a request from the console 2, the image processing device 4, etc. from the stored image data, and transmits it to the requesting device.
• a specific example of the image storage system 3 is PACS (Picture Archiving and Communication Systems).
• the image processing device 4 is a computer such as a workstation, a server computer, or a personal computer, and includes a CPU (Central Processing Unit) 21, a nonvolatile storage 23, and a memory 26 as a temporary storage area. Be prepared.
• the image processing device 4 also includes a display 24 such as a liquid crystal display, an input device 25 such as a keyboard and a mouse, and a network I/F (InterFace) 2 connected to a network (not shown). Equipped with 7.
• the CPU 21, storage 23, display 24, input device 25, memory 26, and network I/F 27 are connected to the bus 28.
• the CPU 21 is an example of a processor in the present disclosure.
• the storage 23 is realized by an HDD (Hard Disk Drive), an SSD (Solid State Drive), a flash memory, or the like.
• the image processing program 22 installed in the image processing device 4 is stored in the storage 23 as a storage medium.
• the CPU 21 reads the image processing program 22 from the storage 23, loads it into the memory 26, and executes the loaded image processing program 22.
• the image processing program 22 is stored in a storage device of a server computer connected to a network or in a network storage in a state that can be accessed from the outside, and is downloaded to a computer constituting the image processing device 4 in response to a request. will be installed. Alternatively, it may be recorded and distributed on a recording medium such as a DVD (Digital Versatile Disc) or a CD-ROM (Compact Disc Read Only Memory), and from that recording medium to the image processing device 4. installed on the computers that configure it.
• a recording medium such as a DVD (Digital Versatile Disc) or a CD-ROM (Compact Disc Read Only Memory)
• FIG. 4 is a diagram showing the functional configuration of the image processing device according to the first embodiment.
• the image processing device 4 includes an image acquisition section 31, a structure extraction section 32, a reconstruction section 33, a feature structure detection section 34, a projection section 35, a displacement amount derivation section 36, and a display control section 37. Equipped with Then, by executing the image processing program 22, the CPU 21 controls the image acquisition section 31, the structure extraction section 32, the reconstruction section 33, the feature structure detection section 34, the projection section 35, the displacement amount derivation section 36, and the display control section. It functions as a section 37.
• the image acquisition unit 31 acquires a plurality of projection images generated by the console 2 causing the imaging device 10 to perform tomosynthesis imaging.
• the image acquisition unit 31 acquires a plurality of projection images from the console 2 or the image storage system 3 via the network I/F 18.
• the console 2 moves the X-ray source 16 by rotating the arm section 12 around the rotation axis 11 when performing tomosynthesis imaging for generating a tomographic image. Further, the console 2 irradiates the breast M, which is the subject, with X-rays at a plurality of radiation source positions by moving the X-ray source 16 according to predetermined imaging conditions for tomosynthesis imaging.
• the radiation detector 15 detects the X-rays that have passed through the breast M, projection images G1, G2, . . . Gn are obtained corresponding to each of the radiation source positions S1 to Sn.
• the breast M is irradiated with the same dose of X-rays at each of the radiation source positions S1 to Sn.
• the radiation source position Sc is a radiation source position where the optical axis X0 of the X-rays emitted from the X-ray source 16 is orthogonal to the detection surface 15A of the radiation detector 15.
• the radiation source position Sc will be referred to as a reference radiation source position Sc
• the projection image Gc obtained by irradiating the breast M with X-rays at the reference radiation source position Sc will be referred to as a reference projection image Gc.
• the optical axis X0 of the X-rays is perpendicular to the detection surface 15A of the radiation detector 15
• the optical axis X0 of the X-rays intersects with the detection surface 15A of the radiation detector 15 at an angle of 90 degrees.
• it is not limited to this, and includes cases where the angle crosses 90 degrees with a certain degree of error.
• the optical axis X0 of X-rays intersects with the detection surface 15A of the radiation detector 15 with an error of about ⁇ 3 degrees with respect to 90 degrees, the "optical axis X0 of X-rays perpendicular to the detection surface 15A of the device 15.
• the structure extraction unit 32 derives a plurality of structural projection images by extracting a specific structure from each of the plurality of projection images Gi.
• the specific structure include at least one of a line structure and a point structure included in the breast M.
• linear structures include mammary glands, spicules, and blood vessels.
• point structures include calcifications, intersections of multiple mammary glands, and intersections of blood vessels. Note that the point structure is not limited to minute points, but also includes regions with a predetermined area. Furthermore, in the following description, both line structures and point structures will be used as specific structures.
• the projection image Gi described in "Concentration Evaluation Method and Vector Concentration Filter, Yoshinaga et al., MEDICAL IMAGING TECHNOLOGY VOL.19 No.3, 2001" is used.
• the degree of concentration of a gradient vector representing the gradient of pixel values at is used.
• the method of extracting a line structure using the degree of concentration uses a line concentration degree filter to obtain gradient vectors on both sides of a search line along a certain direction in the projection image Gi, and evaluates the degree of concentration of the gradient vectors.
• This is a method of extracting search lines with large evaluations as line structures.
• the method of extracting the point structure using the degree of concentration is to use a point concentration degree filter to find gradient vectors in a certain direction in the projection image Gi, evaluate the degree of concentration at which the gradient vectors are concentrated, and select points with large evaluations.
• This is a method to extract as a point structure.
• the coefficients of the concentration filter so as to extract a line structure and a point structure having an actual size of about 1 mm in the case of a line structure and about 0.5 mm in actual size in the case of a point structure.
• the structure extraction unit 32 uses the Harris corner detection method, SIFT (Scale-Invariant Edges, intersections of edges, corners of edges, etc. included in the projected image Gi are extracted as a point structure using algorithms such as FAST (Features from Accelerated Segment Test) or SURF (Speeded Up Robust Features). You may also do so.
• SIFT Scale-Invariant Edges, intersections of edges, corners of edges, etc. included in the projected image Gi are extracted as a point structure using algorithms such as FAST (Features from Accelerated Segment Test) or SURF (Speeded Up Robust Features). You may also do so.
• FIG. 6 is a diagram showing a projection image and a structural projection image.
• the structural projection image SGi is an image in which the line structure included in the projection image Gi is extracted.
• the structure extraction unit 32 may derive the multivalued structural projection image SGi, but derives the binary structural projection image SGi by binarizing the multivalued structural projection image SGi. It may be something.
• the structure extraction unit 32 may extract at least one of a line structure and a point structure included in the projection image Gi using a learning model that has been subjected to machine learning.
• the reconstruction unit 33 derives a structural tomographic image in each of the plurality of tomographic planes of the breast M by reconstructing the plurality of structural projection images SGi.
• the reconstruction unit 33 derives a tomographic image that emphasizes a desired tomographic plane of the breast M by reconstructing all or part of the plurality of projection images Gi while correcting positional deviations as described later. do.
• the reconstruction unit 33 emphasizes a desired tomographic plane of the breast M by reconstructing all or part of the plurality of projection images Gi without correcting the amount of positional deviation, as necessary. Derive a tomographic image.
• the reconstruction unit 33 reconstructs the plurality of structural projection images SGi using a well-known backprojection method such as a simple backprojection method or a filtered backprojection method, and as shown in FIG.
• a three-dimensional coordinate position in a three-dimensional space including the breast M is set, and corresponding pixel positions of the plurality of structural projection images SGi or the plurality of projection images Gi are set with respect to the set three-dimensional coordinate position.
• the pixel value is reconstructed, and the pixel value at that coordinate position is calculated. Note that, as will be described later, when the amount of positional deviation based on the body movement of the breast M during tomosynthesis imaging is derived, the reconstruction unit 33 reconstructs the plurality of projection images Gi while correcting the positional deviation. A corrected tomographic image with body movement corrected is derived.
• the characteristic structure detection unit 34 detects at least one characteristic structure from the plurality of structural tomographic images SDj.
• the characteristic structure includes a point structure and a line structure included in the structural tomographic image SDj.
• FIG. 8 is a diagram for explaining detection of feature structures. Note that in this embodiment, a point structure is detected as a feature structure. Furthermore, here, detection of a characteristic structure from one structural tomographic image SDk among the plurality of structural tomographic images SDj will be described.
• the structural tomographic image SDk includes point-like structures E1 to E3 such as calcifications on the tomographic plane of the breast M from which the structural tomographic image SDk was obtained, and edge intersections E4 and E5 such as the intersections of blood vessels. is included.
• the feature structure detection unit 34 detects point-like structures such as calcifications as point structures, that is, feature structures, from the structural tomographic image SDk using a known algorithm. Additionally, points such as edges, intersections of edges, and corners of edges included in the structural tomographic image SDk are detected as feature structures using algorithms such as the Harris corner detection method, SIFT, FAST, or SURF. It's okay. For example, the feature structure detection unit 34 detects the point-like structure E1 included in the structural tomographic image SDk shown in FIG. 8 as the feature structure F1. All of the point structures have high brightness pixel values, that is, small pixel values in the structural tomographic image SDj. Therefore, in any of these methods, the detected feature structure satisfies a specific threshold condition.
• point-like structures such as calcifications as point structures, that is, feature structures
• the point structure is a point in the structural tomographic image SDj whose pixel value is equal to or less than a predetermined threshold value Th1. Note that if the pixel value of the structural tomographic image SDj is such that the higher the brightness, the larger the value, the detected characteristic structure is a point in the structural tomographic image SDj where the pixel value is equal to or higher than a predetermined threshold value. .
• the feature structure is not limited to a point structure, but even when detecting a line structure as a feature structure, for example, the luminance that satisfies a specific threshold condition is used as the standard, or the known method is used. Algorithms can be used as appropriate. Further, in order to detect the feature structure, a computer-aided image diagnosis (CAD) algorithm may be used.
• CAD computer-aided image diagnosis
• the characteristic structure may be only one pixel in the structural tomographic image SDk, or may be composed of a plurality of pixels representing the position of the characteristic structure.
• a characteristic structure is detected only from one structural tomographic image SDk, but in reality, it is assumed that a plurality of characteristic structures are detected from each of a plurality of structural tomographic images SDj.
• the projection unit 35 determines, based on the positional relationship between the radiation source position at the time of imaging and the radiation detector 15 for each of the plurality of projection images Gi, the characteristic structure F1 is a tomographic plane corresponding to the detected tomographic image.
• a plurality of projection images Gi are projected onto the tomographic plane.
• the projection unit 35 derives a tomographic projection image GTi corresponding to each of the plurality of projection images Gi. Derivation of the tomographic projection image GTi will be described below. Note that in this embodiment, since the characteristic structure is detected in each of the plurality of tomographic images Dj, the plurality of projection images Gi are projected onto each of the plurality of tomographic planes Tj corresponding to the plurality of tomographic images Dj. , a tomographic projection image GTi is derived.
• FIG. 9 is a diagram for explaining the projection of a projection image onto a corresponding tomographic plane.
• FIG. 9 the case where one projection image Gi acquired at the radiation source position Si is projected onto one tomographic plane Tj of the breast M will be described.
• a projection image Gi located on this straight line is placed at a position where a straight line connecting the radiation source position Si and a pixel position on the projection image Gi intersects with the tomographic plane Tj. Project the pixel value of .
• the tomographic image derived from the projection image Gi and the tomographic plane Tj is composed of a plurality of pixels discretely arranged two-dimensionally at a predetermined sampling interval, and the lattice points at the predetermined sampling interval are Pixels are placed in
• short line segments perpendicular to the projection image Gi and the tomographic plane Tj indicate pixel division positions. Therefore, in FIG. 9, the center position of the pixel division positions is the pixel position that is the grid point.
• the coordinates of the radiation source position at the radiation source position Si (sxi, syi, szi), the coordinates of the pixel position Pi on the projection image Gi (pxi, pyi), and the coordinates of the projection position on the tomographic plane Tj (tx, ty , tz) is expressed by the following equation (1).
• the z-axis is perpendicular to the detection surface 15A of the radiation detector 15, and the y-axis is parallel to the direction in which the X-ray source 16 moves on the detection surface of the radiation detector 15. It is assumed that the x-axis is set in a direction perpendicular to the y-axis.
• pxi (tx ⁇ szi-sxi ⁇ tz)/(szi-tz)
• pyi (ty ⁇ szi-syi ⁇ tz)/(szi-tz) (1)
• the projection position on tomographic plane Tj on which the pixel values of projection image Gi are projected can be determined. It can be calculated. Therefore, by projecting the pixel values of the projection image Gi onto the calculated projection position on the tomographic plane Tj, the tomographic plane projection image GTi is derived.
• the intersection of the straight line connecting the radiation source position Si and the pixel position on the projection image Gi and the tomographic plane Tj may not be the pixel position on the tomographic plane Tj.
• the projection position (tx, ty, tz) on the tomographic plane Tj may be located between pixel positions O1 to O4 of the tomographic image Dj on the tomographic plane Tj.
• the pixel value at each pixel position may be calculated by performing an interpolation operation using the pixel values of the projected image at a plurality of projection positions around each of the pixel positions O1 to O4.
• a linear interpolation calculation can be used that weights the pixel value of the projected image at the projection position according to the distance between the pixel position and a plurality of projection positions around the pixel position.
• any method such as nonlinear bicubic interpolation using pixel values at more projection positions around the pixel position and B-spline interpolation can be used.
• the pixel value at the projection position closest to a pixel position may be used as the pixel value at that pixel position. Thereby, pixel values at all pixel positions on the tomographic plane Tj are determined for the projection image Gi.
• a tomographic projection image GTi having pixel values determined at all pixel positions of the tomographic plane Tj in this manner is derived. Therefore, in one tomographic plane, the number of tomographic projection images GTi matches the number of projection images Gi.
• the positional deviation amount deriving unit 36 derives the positional deviation amount between the plurality of tomographic projection images GTi based on the body movement of the breast M during tomosynthesis imaging. First, the positional deviation amount deriving unit 36 sets a local region corresponding to the feature structure F1 as a region of interest for a plurality of tomographic projection images GTi. Specifically, a local region of a predetermined size centered on the coordinate position of the feature structure F1 is set as a region of interest.
• FIG. 11 is a diagram for explaining setting of a region of interest. In FIG. 11, for the sake of explanation, it is assumed that three projection images G1 to G3 are projected onto the tomographic plane Tj and tomographic plane projection images GT1 to GT3 are derived.
• the positional deviation amount deriving unit 36 sets a region of interest Rf0 centered on the coordinate position of the feature structure F1 in the tomographic image Dj on the tomographic plane Tj. Then, regions of interest R1 to R3 corresponding to the region of interest Rf0 are set in each of the tomographic projection images GT1 to GT3. Note that the broken lines in FIG. 11 indicate the boundaries between the regions of interest R1 to R3 and other regions. Therefore, on the tomographic plane Tj, the positions of the region of interest Rf0 and the regions of interest R1 to R3 coincide.
• FIG. 12 is a diagram showing regions of interest R1 to R3 set in tomographic projection images GT1 to GT3.
• the regions of interest R1 to R3 are areas of, for example, 50 ⁇ 50 pixels or 100 ⁇ 100 pixels around the feature structure F1. do it.
• the positional deviation amount deriving unit 36 performs positioning of the regions of interest R1 to R3. At this time, alignment is performed using the region of interest set in the reference tomographic projection image as a reference. In this embodiment, a tomographic projection image (reference tomographic Using the region of interest set in the plane projection image as a reference, other regions of interest are aligned.
• the positional deviation amount deriving unit 36 aligns the regions of interest R1 and R3 with respect to the region of interest R2, and uses a shift vector representing the moving direction and amount of movement of the regions of interest R1 and R3 with respect to the region of interest R2 as the amount of positional deviation.
• alignment involves determining the moving direction and amount of movement of the regions of interest R1 and R3 with respect to the region of interest R2 in a predetermined search range so that the correlation between the regions of interest R1 and R3 with respect to the region of interest R2 is maximized.
• normalized cross-correlation may be used as the correlation.
• the shift vector is one less than the number of tomographic projection images. For example, if the number of tomographic projection images is 15, the number of shift vectors is 14, and if the number of tomographic projection images is 3, the number of shift vectors is 2.
• FIG. 13 is a diagram showing images within the three regions of interest R1 to R3 when no body movement occurs during the acquisition of the projection images G1 to G3.
• FIG. 13 shows the center positions of the regions of interest R1 to R3, that is, the positions P1 to P3 corresponding to the feature structure F1 in the tomographic projection images GT1 to GT3, and Image F2 is indicated by a large circle.
• the positions P1 to P3 corresponding to the feature structure F1 and the feature matches. Therefore, the shift vectors, ie, the amount of positional deviation, of the regions of interest R1 and R3 with respect to the region of interest R2 are both zero.
• FIG. 14 is a diagram showing images within three regions of interest R1 to R3 when a body movement occurs during acquisition of projection image G2 and projection image G3 among projection images G1 to G3.
• the positions P1 and P2 corresponding to the feature structure F1 in the regions of interest R1 and R2 and the positions P1 and P2 corresponding to the feature structure F1 in the regions of interest R1 and R2
• the position of the image F2 of the feature structure F1 included in the image coincides with the position of the image F2. Therefore, the amount of positional deviation of the region of interest R1 with respect to the region of interest R2 is zero.
• the position P3 corresponding to the feature structure F1 in the region of interest R3 and the image F2 of the feature structure F1 included in the region of interest R3 does not match the position.
• the region of interest R3 is moved by an amount and direction relative to the region of interest R2, and as a result, a shift vector V10 having a magnitude and direction is derived as the amount of positional deviation.
• FIG. 15 is a diagram for explaining changing the search range. As shown in FIG. 15, two types of search ranges, a small search range H1 and a large search range H2, are set as search ranges for the regions of interest R1 and R3 with respect to the reference region of interest R2.
• the mammary gland density is low, the amount of fat in the breast M increases, so body movement tends to increase during imaging. Furthermore, when the breast M is large, the body movement tends to be large during imaging. Furthermore, the longer the time for tomosynthesis imaging, the greater the body movement during imaging. Furthermore, even when the compression pressure on the breast M is small, body movements tend to increase during imaging. Furthermore, when the imaging direction of the breast M is MLO (Medio-Lateral Oblique), the body movement tends to be larger during imaging than when CC (Cranio-Caudal) is taken.
• MLO Medio-Lateral Oblique
• the positional deviation amount deriving unit 36 changes the search range when deriving the positional deviation amount. Specifically, if the body movement tends to increase, a large search range H2 shown in FIG. 15 may be set. Conversely, if the body movement tends to become smaller, a small search range H1 shown in FIG. 15 may be set.
• the amount of positional deviation of the plurality of tomographic projection images GTi is derived only for one feature structure F1 detected in one tomographic plane Tj.
• FIG. 10 feature structures).
• positional deviation amounts regarding a plurality of different feature structures F are derived for a tomographic projection image corresponding to a projection image acquired in a state where body movement has occurred.
• the positional deviation amount deriving unit 36 interpolates positional deviation amounts for a plurality of different feature structures F with respect to the coordinate position in the three-dimensional space from which the tomographic image Dj is derived.
• the positional deviation amount deriving unit 36 calculates the positional deviation when performing reconstruction for all coordinate positions in the three-dimensional space from which the tomographic image is derived, for the tomographic projection image acquired while the body movement has occurred. Derive the quantity.
• the reconstruction unit 33 derives a corrected tomographic image Dhj in which the body movement has been corrected by reconstructing the projection image Gi while correcting the positional deviation amount derived in this way. Specifically, when the reconstruction is performed using a back projection method, a pixel of the projection image Gi in which a positional shift has occurred is replaced with a corresponding pixel of another projection image based on the derived positional shift amount. Reconstruction is performed by correcting the positional shift so that the image is projected at the back-projected position.
• one amount of positional deviation may be derived from a plurality of different feature structures F.
• a region of interest is set for each of the plurality of different feature structures F, and the amount of positional shift is derived on the assumption that the entire region of interest has moved by the same amount in the same direction.
• the amount of positional deviation is derived so that the representative value (e.g., average value, intermediate value, maximum value, etc.) of the correlation for all regions of interest between the tomographic projection images that are the target of deriving the amount of positional deviation is maximized. do it.
• the three-dimensional space within the breast M represented by the plurality of tomographic images Dj is divided into a plurality of three-dimensional regions, and one positional deviation amount is derived from the plurality of characteristic structures F for each region in the same manner as above. You can also do this.
• the reconstruction unit 33 derives a plurality of tomographic images Dj by reconstructing a plurality of projection images Gi without correcting the amount of positional deviation.
• FIG. 17 is a diagram showing a display screen of a corrected tomographic image.
• the display screen 40 displays a tomographic image Dj before body movement correction and a corrected tomographic image Dhj after body movement correction.
• a label 41 of "before correction” is given to the tomographic image Dj so that it can be seen that the body movement has not been corrected.
• the corrected tomographic image Dhj is given a label 42 of "after correction” so that it can be seen that the body movement has been corrected.
• the label 41 may be attached only to the tomographic image Dj, or the label 42 may be attached only to the corrected tomographic image Dhj.
• the tomographic image Dj and the corrected tomographic image Dhj display the same cross section. Further, when switching the tomographic plane to be displayed based on an instruction from the input device 25, it is preferable to link the tomographic planes to be displayed in the tomographic image Dj and the corrected tomographic image Dhj. Further, in addition to the tomographic image Dj and the corrected tomographic image Dhj, the projection image Gi may be displayed.
• the operator can check the success or failure of body movement correction by looking at the display screen 40. Furthermore, if the body movement is too large, even if a tomographic image is derived by reconstructing while correcting the amount of positional deviation as in this embodiment, the body movement cannot be accurately corrected, and the body movement correction cannot be performed. It may fail. In such a case, body movement correction may fail and the tomographic image Dj may have higher image quality than the corrected tomographic image Dhj. Therefore, the input device 25 may receive an instruction as to which of the tomographic images Dj and the corrected tomographic images Dhj to save, and the instructed image may be saved in the storage 23 or an external storage device.
• FIG. 18 is a flowchart showing the processing performed in the first embodiment.
• the image acquisition unit 31 acquires a plurality of projection images Gi derived by the console 2 causing the imaging device 10 to perform tomosynthesis imaging of the breast M (step ST1).
• the structure extraction unit 32 derives a plurality of structural projection images SGi by extracting a specific structure from each of the plurality of projection images Gi (step ST2).
• the reconstruction unit 33 derives a structural tomographic image SDj in each of the plurality of tomographic planes of the breast M by reconstructing the plurality of structural projection images SGi (step ST3).
• the characteristic structure detection unit 34 detects at least one characteristic structure from the plurality of structural tomographic images SDj (step ST4).
• the projection unit 35 converts the plurality of projection images Gi into a tomographic image in which the characteristic structure F1 is detected based on the positional relationship between the radiation source position and the radiation detector 15 at the time of imaging for each of the plurality of projection images Gi. , to derive tomographic projection images GTi corresponding to each of the plurality of projection images Gi (step ST5).
• the positional deviation amount deriving unit 36 derives the positional deviation amount between the plurality of tomographic projection images GTi (step ST6). Further, the reconstruction unit 33 derives a corrected tomographic image Dhj by reconstructing the plurality of projection images Gi while correcting the positional deviation (step ST7). Then, the display control unit 37 displays the corrected tomographic image Dhj on the display 24 (step ST8), and the process ends. Note that the derived corrected tomographic image Dhj is transmitted to the image storage system 3 and stored.
• the projected image Gi has a lot of noise.
• structures such as calcification within the breast M may have reduced contrast depending on how the structures overlap, and may be buried in noise in the projection image Gi. Therefore, when a feature structure is detected from a tomographic image derived by reconstructing a plurality of projection images Gi, the feature structure detected from the tomographic image is associated with the structure corresponding to the feature structure included in the projection image. As a result, it may not be possible to accurately correct the positional deviation of the projected image Gi using the feature structure.
• a structural projection image SGi is derived by extracting specific structures such as line structures and point structures from the projection image, and a structural tomographic image SDj is derived by reconstructing the structural projection image SGi. Then, the characteristic structure is detected from the structural tomographic image SDj.
• the structural tomographic image SDj is derived from the structural projection image SGi, it is guaranteed that structures corresponding to the feature structures detected from the structural tomographic image SDj are included in the structural projection image SGi and furthermore in the projection image Gi. That will happen.
• the amount of positional deviation between the plurality of projection images Gi can be appropriately derived using the detected feature structure, and as a result, according to the present embodiment, body movement It is possible to obtain a high-quality corrected tomographic image Dhj in which the influence of .
• characteristic structures are detected not from the projection image Gi or the tomographic projection image GTi, but from a plurality of structural tomographic images SDj.
• the structural tomographic image SDj includes only structures included in the corresponding tomographic plane Tj. Therefore, structures on other tomographic planes that are included in the projection image Gi are not included in the structural tomographic image SDj. Therefore, according to the first embodiment, characteristic structures can be detected with high accuracy without being influenced by structures on other tomographic planes. Therefore, the amount of positional deviation between the plurality of projection images Gi can be appropriately derived, and as a result, according to the present embodiment, a high-quality corrected tomographic image Dhj in which the influence of body movement is reduced is obtained. can do.
• the configuration of the image processing device according to the second embodiment is similar to the configuration of the image processing device according to the first embodiment shown in FIG. Omitted.
• the amount of positional shift is derived between the tomographic projection images GTi.
• a region of interest Rf0 centered on the coordinate position of the characteristic structure F1 is set in the structural tomographic image SDj, and a positional shift of the region of interest Ri set in the tomographic projection image GTi with respect to the set region of interest Rf0 is determined.
• the amount is derived as a temporary positional deviation amount.
• the second embodiment differs from the first embodiment in that the amount of positional deviation between the plurality of tomographic projection images GTi is derived based on the derived provisional amount of positional deviation.
• the region of interest Ri set in the plurality of tomographic projection images GTi corresponds to the first local region
• the region of interest Rf0 set in the structural tomographic image SDj corresponds to the second local region.
• FIG. 19 is a diagram for explaining the derivation of the amount of positional deviation in the second embodiment.
• the region of interest Rf0 and the regions of interest R1 to R3 in FIG. 19 are the same as the region of interest Rf0 and the regions of interest R1 to R3 shown in FIG. 11 and the like described above.
• the positional deviation amount deriving unit 36 first uses the region of interest Rf0 set in the structural tomographic image SDj as a reference, and calculates the tomographic projection image GTi (GT1 to GT3 in FIG.
• GTi tomographic projection image
• the amount of positional deviation of the regions of interest R1 to R3 set in (also not shown) is derived as a provisional amount of positional deviation.
• FIG. 20 is a diagram showing images within three regions of interest R1 to R3 when a body movement occurs during acquisition of projection image G2 and projection image G3 among projection images G1 to G3.
• the positions P1 and P2 corresponding to the feature structure F1 in the regions of interest R1 and R2 and the positions P1 and P2 corresponding to the feature structure F1 in the regions of interest R1 and R2
• the position of the image F2 of the feature structure F1 included in the image coincides with the position of the image F2. Therefore, the amount of positional deviation between the regions of interest R1 and R2 with respect to the region of interest Rf0 is zero.
• the position P3 corresponding to the feature structure F1 in the region of interest R3 and the image F2 of the feature structure F1 included in the region of interest R3 does not match the position. Therefore, the region of interest R3 changes in amount and direction with respect to the region of interest Rf0. Therefore, the shift vectors Vf1 and Vf2 of the regions of interest R1 and R2 with respect to the region of interest Rf0, that is, the tentative positional deviation amount, are 0, but the shift vector Vf3 of the region of interest R3 with respect to the region of interest Rf0, that is, the tentative positional deviation amount is 0. It becomes something that has a value.
• the positional deviation amount deriving unit 36 derives the positional deviation amount between the tomographic projection images GTi based on the provisional positional deviation amount.
• the positional shift is based on the projection image acquired at the reference source position Sc where the optical axis X0 of the X-rays from the X-ray source 16 is perpendicular to the radiation detector 15. Derive the quantity.
• the positional deviation amount deriving unit 36 calculates the positional deviation amount between the tomographic projection image GT1 and the tomographic projection image GT2 from the regions of interest R1, R2 with respect to the region of interest Rf0.
• the positional deviation amount deriving unit 36 calculates the positional deviation amount between the tomographic projection image GT3 and the tomographic projection image GT2 using the difference value Vf3 ⁇ Vf2 of the shift vectors Vf3 and Vf2 of the regions of interest R3 and R2 with respect to the region of interest Rf0. Derive.
• the amount of temporary positional deviation between the region of interest Rf0 set on the structural tomographic image SDj and the regions of interest R1 to R3 set on the tomographic projection image GTi is derived, and Based on the amount of positional deviation, the amount of positional deviation between the tomographic projection images GTi is derived.
• the region of interest Rf0 is set in the structural tomographic image SDj, unlike the projection image Gi, it includes only structures on the tomographic plane from which the structural tomographic image SDj was acquired. Therefore, according to the second embodiment, the amount of positional deviation is derived while reducing the influence of structures included in tomographic planes other than the tomographic plane on which the characteristic structure is set.
• the influence of structures on other tomographic planes can be further reduced, and the amount of positional deviation between the plurality of projection images Gi can be derived with high accuracy, and as a result, the second According to the embodiment, it is possible to obtain a high-quality corrected tomographic image Dhj in which the influence of body movement is reduced.
• the search range when deriving the amount of positional deviation may be changed depending on at least one of the M photographing directions.
• shift vectors Vf1 to Vf3 of the regions of interest R1 to R3 with respect to the region of interest Rf0 are derived as temporary positional deviation amounts, but at this time, as shown in FIG.
• a surrounding area Ra0 smaller than the region of interest Rf0 may be set around the feature structure F1 at Rf0, and the shift vector may be derived based on the surrounding area Ra0.
• the shift vector may be derived using only the surrounding area Ra0.
• the surrounding region Ra0 may be weighted more heavily than the regions other than the surrounding region Ra0 in the regions of interest R1 to R3.
• the region of interest Rf0 is set in the structural tomographic image SDj, but the structural tomographic image to be derived is different for each tomographic projection image GTi from which the temporary positional deviation amount is derived.
• FIG. 22 is a diagram schematically showing the processing performed in the third embodiment.
• the projection image G1 out of the 15 projection images G1 to G15 is assumed to be the target projection image
• the tomographic plane projection image GT1 is assumed to be the target tomographic plane projection image.
• the reconstruction unit 33 reconstructs the structural projection images SG2 to SG15 other than the structural projection image SG1 derived from the target projection image G1 in the tomographic plane Tj to derive a structural tomographic image (SDj_1).
• the feature structure detection unit 34 detects the feature structure from the structural tomographic image SDj_1, the projection unit 35 derives the tomographic projection images GT1 to GT15 from the projection images G1 to G15, and the positional deviation amount deriving unit 36
• a region of interest Rf0_1 is set in the image SDj_1, and a shift vector Vf1 of the region of interest R1 set in the tomographic projection image GT1 for the region of interest Rf0_1 is derived as a temporary positional deviation amount.
• the reconstruction unit 33 reconstructs the structural projection images SG1, SG3 to SG15 other than the structural projection image SG2 derived from the projection image G2, and An image (supposed to be SDj_2) is derived.
• the feature structure detection unit 34 detects the feature structure from the structural tomographic image SDj_2, the projection unit 35 derives the tomographic projection images GT1 to GT15 from the projection images G1 to G15, and the positional deviation amount deriving unit 36
• a region of interest Rf0_2 is set in the image SDj_2, and a shift vector Vf2 of the region of interest R2 set in the tomographic projection image GT2 for the region of interest Rf0_2 is derived as a temporary positional deviation amount.
• the target tomographic projection images are sequentially changed to derive temporary positional deviation amounts for all the tomographic projection images GTi, and from the temporary positional deviation amounts, the tomographic projection images are calculated as in the second embodiment.
• the amount of positional deviation between GTi is derived.
• a temporary positional shift amount is derived using a structural tomographic image that is not affected by the target projection image. Therefore, the temporary positional deviation amount can be derived with higher accuracy, and as a result, the positional deviation amount can be derived with higher accuracy.
• the configuration of the image processing device according to the fourth embodiment is similar to the configuration of the image processing device according to the first embodiment shown in FIG. Explanation will be omitted.
• the structural tomographic image SDj is updated by reconstructing the structural projection image SGi while correcting the amount of positional deviation, and the updated threshold is used to calculate the structural tomographic image from the updated structural tomographic image.
• the updated feature structure is detected, the amount of positional deviation is updated using the updated feature structure, the structural tomographic image is updated until the amount of positional deviation converges, and the updated threshold value is used to update the positional deviation amount.
• This embodiment differs from the first embodiment in that detection of the feature structure and updating of the amount of positional deviation are repeated.
• FIG. 23 is a flowchart showing the processing performed in the fourth embodiment. Note that in FIG. 23, the processing from step ST11 to step ST16 is the same as the processing from step ST1 to step ST6 shown in FIG. 18, so detailed explanation will be omitted here.
• the positional deviation amount deriving unit 36 determines whether the positional deviation amount has converged (step ST17). It may be determined whether the amount of positional deviation has converged or not by determining whether the amount of positional deviation derived for each tomographic projection image GTi has become equal to or less than a predetermined threshold value Th10.
• the threshold value Th10 may be set to a value that allows it to be said that there is no influence of body movement on the tomographic image even if the amount of positional deviation is not corrected any further. Note that it is possible to determine whether or not the positional deviation amount has converged by determining whether a representative value such as the average value of the positional deviation amount derived for the plurality of tomographic projection images GTi has become equal to or less than the threshold value Th10. You may also make a determination.
• step ST17 the reconstruction unit 33 updates the structural tomographic image by reconstructing the plurality of structural projection images SGi while correcting the amount of positional deviation (step ST18). Then, the process returns to step ST14 and processes from step ST14 to step ST17 are performed.
• the feature structure detection unit 34 updates the threshold value Th1 used when detecting the feature structure in the first process of step ST14.
• the pixel value of the structural tomographic image SDj becomes a smaller value as the luminance becomes higher, so the updated threshold value Th2 is a value smaller than the threshold value Th1 used in the first step ST14 processing. Detect the updated feature structure using .
• the projection unit 35 updates the plurality of projection images Gi based on the positional relationship between the radiation source position and the radiation detector 15 at the time of imaging for each of the plurality of projection images Gi.
• the updated tomographic projection image GTi corresponding to each of the plurality of projection images Gi is derived by projecting the detected feature structure F1 onto a corresponding tomographic plane corresponding to the detected tomographic image.
• the positional deviation amount deriving unit 36 derives the updated positional deviation amount between the updated plurality of tomographic projection images GTi.
• an updated threshold value Th2 that is a larger value than the threshold value Th1 used in the first processing of step ST14 is used. Detect updated feature structure.
• step ST17 is negative, steps ST18 and steps ST14 to ST16 are repeated until step ST17 is affirmed.
• the feature structure detection unit 34 detects the feature structure from the updated structural tomographic image SDj using the updated threshold value.
• step ST17 the reconstruction unit 33 derives a corrected tomographic image Dhj by reconstructing the plurality of projection images Gi while correcting the updated positional deviation amount (step ST19). Then, the display control unit 37 displays the corrected tomographic image Dhj on the display 24 (step ST20), and the process ends. Note that the derived corrected tomographic image Dhj is transmitted to the image storage system 3 and stored.
• the structural tomographic image SDj is updated by reconstructing the structural projection image SGi while correcting the amount of positional deviation, and the updated structure is updated using the updated threshold value.
• the updated feature structure is detected from the tomographic image, the amount of positional deviation is updated using the updated feature structure, and the structural tomographic image is updated until the amount of positional deviation converges. Detection of updated feature structures and updating of positional deviation amounts are repeated. Therefore, positional displacement caused by body movement can be more effectively removed, and as a result, a tomographic image of higher quality can be obtained.
• the process of updating the positional deviation amount may be repeated until the positional deviation amount converges.
• the structural tomographic image is updated until the positional deviation amount converges, and the updated feature structure is detected using the updated threshold value.
• the positional shift amount is repeatedly updated, the present invention is not limited to this. The updating of the structural tomographic image, the detection of the updated feature structure, and the updating of the positional deviation amount may be repeated without updating the threshold value.
• the process of updating the positional deviation amount is repeated until the positional deviation amount converges, but the process is not limited to this.
• the process of updating the amount of positional deviation may be repeated a predetermined number of times.
• the positional deviation amount derived by the positional deviation amount deriving unit 36 is compared with a predetermined threshold value, and only when the positional deviation amount exceeds the threshold value, the positional deviation amount is determined.
• the tomographic image may be reconstructed while correcting.
• the threshold value may be set to a value that allows it to be said that there is no effect of body movement on the tomographic image even without correcting the amount of positional deviation.
• a warning display 45 may be displayed on the display 24 to notify that the body movement has exceeded the threshold. The operator can instruct whether or not to perform body movement correction by selecting YES or NO on the warning display 45.
• a region of interest is set in the structural tomographic image SDj and the tomographic projection image GTi, and the moving direction and direction of the region of interest are set.
• the movement amount is derived as a shift vector, that is, a positional deviation amount and a temporary positional deviation amount
• the present invention is not limited to this.
• the amount of positional deviation may be derived without setting a region of interest.
• FIG. 25 is a diagram showing the functional configuration of an image processing apparatus according to the fifth embodiment. Note that in FIG. 25, components similar to those in FIG. 4 are given the same reference numbers as in FIG. 4, and detailed explanations are omitted here.
• the image processing device 4A according to the fifth embodiment includes a focal plane determination unit 38 that determines whether a corresponding tomographic plane corresponding to a structural tomographic image in which each of the plurality of feature structures F is detected is a focal plane.
• This embodiment differs from the first embodiment in that the second embodiment further includes a positional deviation amount deriving unit 36 that derives the positional deviation amount in the corresponding tomographic plane determined to be the focal plane. Note that although the processing according to the fifth embodiment is also applicable to the second to fourth embodiments, only the case where it is applied to the first embodiment will be described here.
• FIG. 26 is a diagram for explaining ripple artifacts.
• the ripple artifact of the structure 48 is included in the tomographic images corresponding to the upper and lower tomographic planes of the tomographic image D3.
• the ripple artifact expands and becomes blurred as it moves away from the fault plane that includes the structure 48.
• the range in which the ripple artifact spreads corresponds to the range in which the X-ray source 16 moves.
• Ripple artifacts also occur in the structural tomographic image SDj.
• the feature structure F detected by the feature structure detection unit 34 from the structural tomographic image SDj of the corresponding tomographic plane is a ripple artifact
• the feature structure F is blurred and spread over a wide range. Therefore, if such a feature structure F is used, it is not possible to accurately derive the amount of positional deviation.
• the focal plane determination unit 38 determines whether or not the corresponding tomographic plane for the structural tomographic image SDj in which the characteristic structure F has been detected is the focal plane.
• the projection unit 35 derives the tomographic projection image GTi
• the positional deviation amount derivation unit 36 derives the positional deviation amount.
• the amount of positional shift is derived using the feature structure detected from the structural tomographic image in the corresponding tomographic plane determined to be the focal plane. Determination of whether or not it is the focal plane will be described below.
• the focal plane discrimination unit 38 derives corresponding points corresponding to the feature structures in the plurality of structural tomographic images SDj for the feature structures detected by the feature structure detection unit 34.
• FIG. 27 is a diagram for explaining derivation of corresponding points. As shown in FIG. 27, if a characteristic structure F3 is detected in a certain structural tomographic image SDk, the positional deviation amount deriving unit 36 detects the characteristic structure F3 in a plurality of structural tomographic images located in the thickness direction of the structural tomographic image SDk. Corresponding points C1, C2, C3, C4, etc. corresponding to F3 are derived. Note that in the following description, the reference numeral C is used for the corresponding point.
• the corresponding point C may be derived by aligning the region of interest including the feature structure F3 with a structural tomographic image other than the structural tomographic image SDk. Then, the focal plane discrimination unit 38 plots the pixel values of the feature structure F3 and the corresponding points C in the order in which the tomographic planes are arranged.
• FIG. 28 is a diagram showing the result of plotting the feature structure and pixel values of corresponding points. As shown in FIG. 28, the feature structure and the pixel values of corresponding points change to have minimum values in the feature structure due to the influence of ripple artifacts.
• the feature structure F3 is on the focal plane, the feature structure F3 is not blurred and has high brightness, that is, the pixel value is small.
• the feature structure F3 is a ripple artifact, so the pixel value is blurred and the pixel value becomes larger than the minimum value.
• the focal plane discrimination unit 38 determines that the position of the tomographic plane where the feature structure F3 is detected is the position P0 shown in FIG. 28 where the pixel value is the minimum. If so, it is determined that the corresponding tomographic plane where the feature structure F3 is detected is the focal plane. On the other hand, if the position of the tomographic plane where the characteristic structure F3 is detected is a position P1 shown in FIG. 28 where the pixel value is not the minimum, it is determined that the corresponding tomographic plane where the characteristic structure F3 is detected is not the focal plane.
• the projection unit 35 derives a tomographic projection image GTi in the same manner as in each of the above embodiments only on the corresponding tomographic plane determined to be the focal plane.
• the positional deviation amount deriving unit 36 derives the positional deviation amount of the tomographic projection image GTi in the corresponding tomographic plane determined to be the focal plane. That is, the positional deviation amount deriving unit 36 derives the positional deviation amount of the tomographic projection image GTi using the characteristic structure detected in the corresponding tomographic plane determined to be the focal plane.
• FIG. 29 is a flowchart showing the processing performed in the fifth embodiment. Note that the processing from step ST21 to step ST24 in FIG. 29 is similar to the processing from step ST1 to step ST4 in FIG. 18, so a detailed explanation will be omitted here. Note that in the fifth embodiment, it is assumed that a plurality of feature structures are detected.
• the focal plane discrimination unit 38 determines the corresponding tomographic plane corresponding to the structural tomographic image in which each of the plurality of feature structures detected by the feature structure detection unit 34 is detected. It is determined whether or not it is the focal plane (focal plane determination; step ST25). Then, the projection unit 35 derives a tomographic projection image GTi on the corresponding tomographic plane determined to be the focal plane (step ST26), and the positional deviation amount deriving unit 36 derives the tomographic projection image GTi on the corresponding tomographic plane determined to be the focal plane. The amount of positional deviation is derived using the characteristic structure detected in the structural tomographic image of the plane (step ST27).
• the reconstruction unit 33 derives a corrected tomographic image Dhj by reconstructing the plurality of projection images Gi while correcting the amount of positional deviation (step ST28). Then, the display control unit 37 displays the corrected tomographic image Dhj on the display 24 (step ST29), and the process ends. Note that the derived corrected tomographic image Dhj is transmitted to the image storage system 3 and stored.
• the amount of positional shift is derived in the corresponding tomographic plane determined to be the focal plane. Therefore, the amount of positional deviation can be derived with high accuracy without being influenced by ripple artifacts, and as a result, the corrected tomographic image Dhj in which the positional deviation has been corrected with high accuracy can be derived.
• the corresponding tomographic plane is a focal plane using the feature structure and the plotting results of pixel values of corresponding points. This determination is not limited to this. Between the feature structure and the ripple artifact, the difference in contrast between the feature structure and the surrounding pixels is greater in the feature structure. For this reason, the contrast between the feature structure and the surrounding pixels at the corresponding point may be derived, and if the contrast for the feature structure is maximum, it may be determined that the corresponding tomographic plane where the feature structure was detected is the focal plane. .
• the pixel values at the positions corresponding to the feature structures in the projection images will have small variations between projection images if the feature structures are in the focal plane, but if the feature structures are not in the focal plane, the feature structures will be Since there is a possibility that a structure other than the structure corresponding to the projection image is represented, the variation between projection images becomes large. Therefore, the variance value of pixel values corresponding to the feature structure between the projection images Gi is derived, and if the variance value is less than or equal to a predetermined threshold, the corresponding tomographic plane where the feature structure was detected is the focal plane.
• the present invention may also include a discriminator that determines whether or not the corresponding tomographic plane in which the characteristic structure has been detected is a focal plane.
• FIG. 30 is a diagram showing the functional configuration of an image processing apparatus according to the sixth embodiment.
• the image processing device 4B according to the sixth embodiment evaluates the image quality of the region of interest including the characteristic structure in the corrected tomographic image Dhj, and determines whether the derived positional deviation amount is appropriate or inappropriate based on the image quality evaluation result.
• This embodiment differs from the first embodiment in that it further includes a positional deviation amount determining section 39 that determines whether the positional deviation amount is the same or not. Note that although the processing according to the sixth embodiment is also applicable to the second to fifth embodiments, only the case where it is applied to the first embodiment will be described here.
• the positional deviation amount determination unit 39 determines the interest centering on the coordinate positions of a plurality of (in this case, two) feature structures F4 and F5 included in the corrected tomographic image Dhj, as shown in FIG. Areas Rh1 and Rh2 are set. Then, a high frequency image is derived by extracting high frequency components in each of the regions of interest Rh1 and Rh2.
• the high-frequency components may be extracted by, for example, performing filtering processing using a Laplacian filter or the like to derive a second-order differential image, but the present invention is not limited thereto. Further, the positional deviation amount determining unit 39 derives the magnitude of the high frequency component of the regions of interest Rh1 and Rh2.
• the magnitude of the high frequency component may be derived from the sum of squares of pixel values of the high frequency image, but is not limited to this. Then, the positional deviation amount determining unit 39 derives the sum of the magnitudes of the high frequency components for all the regions of interest Rh1 and Rh2.
• the positional deviation amount determination unit 39 performs image quality evaluation based on the magnitude of the high frequency component. That is, the positional deviation amount determining unit 39 determines whether the sum of the magnitudes of the high frequency components for all the regions of interest Rh1 and Rh2, derived as described above, is equal to or greater than the predetermined threshold Th20. judge.
• the positional deviation amount determining unit 39 determines that the positional deviation amount is appropriate; if the total sum is less than the threshold value Th20, the positional deviation amount determining unit 39 determines that the positional deviation amount is appropriate. , the amount of positional deviation is determined to be inappropriate. If the positional deviation amount judgment unit 39 determines that the positional deviation amount is inappropriate, the reconstruction unit 33 reconstructs the plurality of projection images Gi and derives the tomographic image Dj without correcting the positional deviation amount. . Then, the display control unit 37 displays the uncorrected tomographic image Dj on the display 24 instead of the corrected tomographic image Dhj. In this case, the tomographic image Dj before correction is sent to the external storage device instead of the corrected tomographic image Dhj.
• FIG. 32 is a flowchart showing the processing performed in the sixth embodiment. Note that the processing from step ST31 to step ST37 in FIG. 32 is similar to the processing from step ST1 to step ST7 in FIG. 18, so detailed explanation will be omitted here.
• the positional deviation amount determination unit 39 evaluates the image quality of the region of interest including the characteristic structure in the corrected tomographic image Dhj, and based on the image quality evaluation result, It is determined whether the amount of positional deviation is appropriate (step ST38).
• the display control unit 37 displays the corrected tomographic image Dhj on the display 24 (step ST39), and the process ends. Note that the derived corrected tomographic image Dhj is transmitted to the image storage system 3 and stored.
• the reconstruction unit 33 reconstructs the plurality of projection images Gi and derives the tomographic image Dj without correcting the amount of positional deviation (step ST40). Then, the display control unit 37 displays the tomographic image Dj on the display 24 (step ST41), and the process ends. in this case, The tomographic image Dj is transmitted to the image storage system 3 and stored.
• the positional deviation amount deriving unit 36 derives the positional deviation amount, it may not be possible to derive an appropriate positional deviation amount due to the influence of structures other than the characteristic structure.
• the image quality of the corrected tomographic image Dhj is evaluated, and based on the image quality evaluation result, it is determined whether the amount of positional deviation is appropriate or inappropriate. Therefore, it is possible to appropriately judge whether the derived positional deviation amount is appropriate or not.
• the amount of positional deviation is determined to be inappropriate, the tomographic image Dj before correction is displayed or saved, so that the correction derived based on the inappropriate amount of positional deviation is corrected. The possibility of incorrect diagnosis being made can be reduced by using the completed tomographic image Dhj.
• the image quality is evaluated based on the magnitude of the high frequency component of the region of interest set in the corrected tomographic image Dhj, but the present invention is not limited to this.
• the reconstruction unit 33 derives a plurality of tomographic images Dj by reconstructing the plurality of projection images Gi without performing positional deviation correction, and the positional deviation amount determination unit 39 calculates the tomographic image Dj of interest including the characteristic structure.
• the image quality of the area is further evaluated, and the image quality evaluation results for the corrected tomographic image Dhj and the image quality evaluation results for the tomographic image Dj are compared, and the tomographic image with the higher image quality evaluation is selected as the final tomographic image. may be determined.
• the final tomographic image is a tomographic image that is displayed on the display 24 or transmitted to and stored in an external device.
• the amount of positional deviation may be repeatedly updated similarly to the fourth embodiment.
• the positional deviation amount derived by the positional deviation amount deriving unit 36 is compared with a predetermined threshold value, and the positional deviation amount exceeds the threshold value. Only in this case, the tomographic image may be reconstructed while correcting the amount of positional deviation.
• FIG. 33 is a diagram showing the functional configuration of an image processing apparatus according to the seventh embodiment.
• the image processing device 4C according to the seventh embodiment includes an evaluation function derivation unit 50 that derives an evaluation function for evaluating the image quality of a region of interest including a feature structure in a corrected tomographic image Dhj, and a positional deviation amount derivation unit 36.
• this embodiment differs from the first embodiment in that a positional shift amount that optimizes the evaluation function is derived. Note that although the processing according to the seventh embodiment is also applicable to the second to fifth embodiments, only the case where it is applied to the first embodiment will be described here.
• the evaluation function deriving unit 50 derives a high-frequency image for the region of interest corresponding to the feature structure F that the positional deviation amount deriving unit 36 has set for the tomographic projection image GTi.
• the high-frequency image may be derived by performing filtering processing using a Laplacian filter or the like to derive a second-order differential image, similarly to the positional deviation amount determination unit 39 in the sixth embodiment.
• Let qkl be the pixel value of the high-frequency image within the derived region of interest.
• k represents the k-th projection image
• l represents the number of pixels within the region of interest.
• a transformation matrix for correcting the amount of positional deviation is Wk
• a transformation parameter in the transformation matrix is ⁇ k.
• the conversion parameter ⁇ k corresponds to the amount of positional deviation.
• the image quality evaluation value for the region of interest corresponding to the feature structure F in the corrected tomographic image Dhj can be regarded as the sum of the sizes of the high-frequency images for the region of interest after positional deviation correction in each of the projection images Gi. can.
• the evaluation function derivation unit 50 derives the evaluation function shown in equation (3) below.
• the evaluation function Ec shown in equation (3) is an evaluation function Ec that calculates a conversion parameter ⁇ k to minimize the value in the parentheses on the right side with a negative value in order to maximize the above addition result.
• the evaluation function shown in equation (3) has multiple local solutions. For this reason, constraints are imposed on the range and average value of the conversion parameter ⁇ k. For example, a constraint is given such that the average of the transformation parameters ⁇ k for all projected images is 0. More specifically, when the transformation parameter ⁇ k is a movement vector representing parallel movement, a constraint condition is given such that the average value of the movement vectors for all projection images Gi is zero.
• the positional deviation amount deriving unit 36 derives a conversion parameter ⁇ k, that is, a positional deviation amount that minimizes the evaluation function Ec shown in equation (3) below.
• the seventh embodiment includes the evaluation function derivation unit 50 that derives an evaluation function for evaluating the image quality of the region of interest including the feature structure in the corrected tomographic image Dhj, and the positional deviation amount derivation unit 36.
• the amount of positional deviation that optimizes the evaluation function is derived. Therefore, it is possible to reduce the possibility that an incorrect diagnosis will be made using the corrected tomographic image Dhj derived based on an inappropriate amount of positional deviation.
• a region of interest is set in the tomographic image Dj and the tomographic projection image GTi, and the moving direction and movement of the region of interest are determined.
• the amount is derived as a shift vector, that is, a positional deviation amount and a temporary positional deviation amount
• the present invention is not limited to this.
• the amount of positional deviation may be derived without setting a region of interest.
• the projection unit 35 derives the tomographic projection image GTi
• the positional deviation amount deriving unit 36 derives the positional deviation amount between the tomographic projection images GTi, but the present invention is not limited to this. It's not something you can do.
• the amount of positional deviation between the projection images Gi may be derived without deriving the tomographic projection images GTi. In this case, the projection unit 35 becomes unnecessary in each of the above embodiments.
• the positional deviation amount deriving unit 36 may derive the positional deviation amount based on the positional relationship of the projection image Gi in the corresponding tomographic plane corresponding to the tomographic image in which the feature structure F has been detected.
• the subject is the breast M, but the subject is not limited to this, and it goes without saying that the subject may be any part of the human body, such as the chest or abdomen.
• the various processors may be used. I can do it.
• the various processors listed above include circuits such as FPGA (Field Programmable Gate Array) after manufacturing.
• Programmable logic devices (PLDs) which are processors whose configuration can be changed, and specialized electrical devices, which are processors with circuit configurations specifically designed to execute specific processes, such as ASICs (Application Specific Integrated Circuits). Includes circuits, etc.
• One processing unit may be composed of one of these various types of processors, or a combination of two or more processors of the same type or different types (for example, a combination of multiple FPGAs or a combination of a CPU and an FPGA). ). Further, the plurality of processing units may be configured with one processor.
• one processor is configured with a combination of one or more CPUs and software, There is a form in which this processor functions as a plurality of processing units.
• processors that use a single IC (Integrated Circuit) chip, such as System On Chip (SoC), which implements the functions of an entire system including multiple processing units. be.
• SoC System On Chip
• various processing units are configured using one or more of the various processors described above as a hardware structure.
• circuitry that is a combination of circuit elements such as semiconductor elements can be used.

Landscapes

• Engineering & Computer Science (AREA)
• Health & Medical Sciences (AREA)
• Life Sciences & Earth Sciences (AREA)
• Medical Informatics (AREA)
• General Health & Medical Sciences (AREA)
• Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
• Radiology & Medical Imaging (AREA)
• Physics & Mathematics (AREA)
• Computer Vision & Pattern Recognition (AREA)
• Public Health (AREA)
• Optics & Photonics (AREA)
• Heart & Thoracic Surgery (AREA)
• Veterinary Medicine (AREA)
• Animal Behavior & Ethology (AREA)
• Surgery (AREA)
• Biophysics (AREA)
• High Energy & Nuclear Physics (AREA)
• Molecular Biology (AREA)
• Biomedical Technology (AREA)
• Pathology (AREA)
• General Physics & Mathematics (AREA)
• Theoretical Computer Science (AREA)
• Quality & Reliability (AREA)
• Epidemiology (AREA)
• Primary Health Care (AREA)
• Dentistry (AREA)
• Oral & Maxillofacial Surgery (AREA)
• Apparatus For Radiation Diagnosis (AREA)

Priority Applications (3)

Applications Claiming Priority (2)

Related Child Applications (1)

Publications (1)

Family

ID=87936539

Family Applications (1)

Country Status (4)

Citations (10)

Family Cites Families (1)

• 2022
  • 2022-12-15 EP EP22931043.8A patent/EP4473912A4/en active Pending
  • 2022-12-15 WO PCT/JP2022/046280 patent/WO2023171073A1/ja not_active Ceased
  • 2022-12-15 JP JP2024505907A patent/JPWO2023171073A1/ja active Pending
• 2024
  • 2024-08-30 US US18/820,266 patent/US20240420329A1/en active Pending

Patent Citations (10)

Non-Patent Citations (3)

Also Published As

Similar Documents

Legal Events