Magnetic resonance imaging support method and magnetic resonance imaging device

Abstract

To provide a magnetic resonance imaging support method for supporting generation of a scan plan of an appropriate actual scan.SOLUTION: The method includes an acquisition step, a multiplanar reconstruction step, a first determination step, and a second determination step. In the acquisition step, a plurality of first slice images corresponding to different slice positions and corresponding to one of a sagittal plane and a coronal plane are acquired by performing a scout scan on a subject. In the multi-planar reconstruction step, a second slice image corresponding to an axial plane by multi-planar reconstruction is generated based on the plurality of first slice images. In the first determination step, a position corresponding to a specific site of the subject is determined based on the second slice image. In the second determination step, a range to be scanned in the actual scan is determined based on the plurality of first slice images at the position determined in the first determination step.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The embodiments disclosed in this specification and the drawings relate to a magnetic resonance imaging assistance method and a magnetic resonance imaging apparatus. [Background technology]

[0002] In recent years, whole-body magnetic resonance imaging technology using a magnetic resonance imaging apparatus has been attracting attention. In whole-body magnetic resonance imaging, a scan of the entire body of a subject is performed using multiple subscans. In each subscan, the table on which the subject rests is moved so that a specific region of the subject is positioned within the scan area of ​​the magnetic resonance imaging apparatus. By repeating the table movement and scans, partial images of multiple consecutive regions of the subject are acquired. After all partial images of the subject have been acquired, the partial images are stitched together in the direction of table movement to generate a whole-body image of the subject.

[0003] In order to avoid a decrease in image accuracy and incompleteness due to the subject being off-center when performing a magnetic resonance imaging scan, it is common to first generate a scout image of the subject by a scout scan before the actual scan, and then determine a scan plan for the actual scan based on the scout image. Here, in order to shorten the scan time of the scout scan, the scout image is generated by scanning the sagittal or coronal plane of the subject.

[0004] In whole-body magnetic resonance imaging, when determining a scan plan for a real scan, a scout scan of the subject's entire body is performed, and a whole-body scout image of the subject is generated. Conventionally, scan plans for real scans of whole-body magnetic resonance imaging are manually generated by an operator, resulting in a problem of long work hours and laborious processes. Determining a scan plan for whole-body magnetic resonance imaging requires time and labor, particularly determining the start and end positions of the scan in the vertical direction of the subject, and determining the central axis of the scan in the horizontal and front-to-back directions of the subject. To identify the start and end positions of the scan, the operator must manually identify the positions of the subject's vertex, head center, knees, feet, etc., based on the whole-body scout image. Furthermore, to determine the central axis of the scan, the operator must manually mark the central axis of the subject based on the whole-body scout image.

[0005] To address this issue, a technology has been proposed that uses an object detection algorithm or the like to extract position information from a scout image and assist in generating a scan plan. For example, a technology (e.g., Siemens' Whole-Body Dot Engine technology) is known that detects landmarks of a subject from a whole-body coronal scout image to assist in determining a scan plan for a full-body magnetic resonance imaging scan. However, this technology uses a single slice image as the scout image, and therefore the positions of the detected landmarks depend on the slice position of the scout image, resulting in inaccurate positioning of the landmarks and the generation of an inappropriate scan plan. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2022-016591 [Patent Document 2] Japanese Patent Publication No. 2022-184766 Summary of the Invention [Problem to be solved by the invention]

[0007] One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to support the generation of an appropriate scan plan for a production scan. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described below can also be considered as other problems. [Means for solving the problem]

[0008] A magnetic resonance imaging assistance method according to an embodiment includes an acquisition step, a multiplanar reconstruction step, a first determination step, a second determination step, and an actual scan step. The acquisition step involves performing a scout scan on a subject to acquire multiple first slice images corresponding to different slice positions, the multiple first slice images corresponding to one of a sagittal plane and a coronal plane. The multiplanar reconstruction step involves generating a second slice image corresponding to an axial plane by multiplanar reconstruction based on the multiple first slice images. The first determination step involves determining a position corresponding to a specific region of the subject based on the second slice image. The second determination step involves determining a range to be scanned in the actual scan based on the multiple first slice images at the position determined in the first determination step. The actual scan step involves scanning the subject within the range determined in the second determination step. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram showing an example of the configuration of a magnetic resonance imaging apparatus according to the first embodiment. [Figure 2] FIG. 2 is a flowchart showing the magnetic resonance imaging assistance method according to the first embodiment. [Figure 3] FIG. 3 shows an example of a partial slice image stack. [Figure 4] FIG. 4 shows an example of a whole-body slice image stack. [Figure 5] FIG. 5 is a schematic diagram showing an example of generating an axial whole-body slice image based on a sagittal whole-body slice image stack. [Figure 6] FIG. 6 is a schematic diagram showing the results of object detection for the whole-body slice images of the multiple axial planes shown in FIG. [Figure 7] FIG. 7 is a schematic diagram showing rough divisions of the human body in a whole-body slice image in a sagittal plane that is set based on a whole-body slice image in an axial plane. [Figure 8] FIG. 8 is a schematic diagram illustrating a method for locating the vertex and center of the head based on axial and sagittal whole-body slice image stacks. [Figure 9] FIG. 9 is a schematic diagram illustrating a method for locating the knee center based on axial and coronal whole-body slice image stacks. [Figure 10] FIG. 10 is a schematic diagram illustrating a method for locating the vertebrae based on a whole-body slice image stack in axial, sagittal, and coronal planes. [Figure 11] FIG. 11 is a diagram showing an example of a scan plan for an actual scan displayed on the display unit. [Figure 12] FIG. 12 is a diagram for explaining the problems of the conventional magnetic resonance imaging assistance method. [Figure 13] FIG. 13 is a flowchart showing a magnetic resonance imaging assistance method according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] First Embodiment The first embodiment relates to a magnetic resonance imaging assistance method and a magnetic resonance imaging apparatus. Hereinafter, the magnetic resonance imaging assistance method and the magnetic resonance imaging apparatus according to the first embodiment will be described with reference to the drawings.

[0011] 1 is a diagram showing an example of the configuration of a magnetic resonance imaging apparatus 100 according to the first embodiment. The magnetic resonance imaging apparatus 100 includes a static magnetic field magnet 101, a static magnetic field power supply (not shown), a gradient magnetic field coil 102, a gradient magnetic field power supply 103, a bed 104, a bed control circuit 105, a transmitting coil 106, a transmitting circuit 107, a receiving coil 108, a receiving circuit 109, a sequence control circuit 110, and a console 120. The magnetic resonance imaging apparatus 100 further includes a gantry (not shown) that functions as a support unit for the static magnetic field magnet 101, the gradient magnetic field coil 102, the transmitting coil 106, the receiving coil 108, etc.

[0012] The static magnetic field magnet 101 is a magnet formed in a hollow, approximately cylindrical shape, and generates a static magnetic field in the internal space. The static magnetic field magnet 101 is, for example, a superconducting magnet, and is excited by receiving a current from a static magnetic field power supply. The static magnetic field power supply supplies a current to the static magnetic field magnet 101. As another example, the static magnetic field magnet 101 may be a permanent magnet, in which case the magnetic resonance imaging apparatus 100 does not need to include a static magnetic field power supply. Alternatively, a static magnetic field power supply may be provided separately from the magnetic resonance imaging apparatus 100.

[0013] The gradient magnetic field coil 102 is a hollow, approximately cylindrical coil, and is disposed inside the static magnetic field magnet 101. The gradient magnetic field coil 102 is formed by combining three coils corresponding to the mutually orthogonal X, Y, and Z axes, and these three coils are individually supplied with current from a gradient magnetic field power supply 103 to generate a gradient magnetic field whose magnetic field strength changes along each of the X, Y, and Z axes. The Z-axis direction is the same direction as the static magnetic field, the Y-axis direction is the vertical direction, and the X-axis direction is perpendicular to the Z-axis and Y-axis.

[0014] The gradient magnetic field power supply 103 supplies current to the gradient magnetic field coil 102 under the control of a sequence control circuit 110 .

[0015] The bed 104 has a top board 104a on which the subject P is placed, and under the control of a bed control circuit 105, the top board 104a is inserted into the cavity of the gradient magnetic field coil 102 with the subject P placed thereon.

[0016] The transmission coil 106 is disposed inside the gradient magnetic field coil 102, receives RF pulses from a transmission circuit 107, and generates a high frequency magnetic field.

[0017] The transmission circuit 107 supplies an RF pulse corresponding to the Larmor frequency to the transmission coil 106. The Larmor frequency is determined according to the type of atom of interest and the magnetic field strength.

[0018] The receiving coil 108 is disposed inside the gradient magnetic field coil 102 and receives magnetic resonance signals emitted from the subject P due to the influence of the high frequency magnetic field. Upon receiving the magnetic resonance signals, the receiving coil 108 outputs the received magnetic resonance signals to a receiving circuit 109. The transmitting coil 106 and the receiving coil 108 may be configured as a single coil having a transmitting and receiving function.

[0019] The receiving circuitry 109 detects the magnetic resonance signals output from the receiving coil 108 and generates k-space data based on the detected magnetic resonance signals. Specifically, the receiving circuitry 109 performs analog-to-digital conversion on the analog magnetic resonance signals output from the receiving coil 108 to generate k-space data. Then, the receiving circuitry 109 transmits the generated k-space data to the sequence control circuit 110. The receiving circuitry 109 may be arranged on the gantry side on which the static magnetic field magnet 101, the gradient magnetic field coil 102, etc. are provided.

[0020] The sequence control circuit 110 drives the gradient magnetic field power supply 103, the transmission circuitry 107, and the reception circuitry 109 in accordance with sequence information transmitted from the console 120, thereby causing the subject P to be scanned and transmitting the scanned k-space data to the console 120. The sequence information defines the intensity of the current supplied from the gradient magnetic field power supply 103 to the gradient magnetic field coil 102, the timing of supplying the current, the intensity of the RF pulse supplied from the transmission circuitry 107 to the transmission coil 106, the timing of applying the RF pulse, the timing of detecting the magnetic resonance signal by the reception circuitry 109, etc. The sequence control circuit 110 is configured by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array), or an electronic circuit such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit), for example.

[0021] The console 120 includes an input / output unit 121, a display unit 122, a communication unit 123, a storage unit 124, an image reconstruction unit 125, an image processing unit 126, and a scan plan generation unit 127. The input / output unit 121, the display unit 122, the communication unit 123, the storage unit 124, the image reconstruction unit 125, the image processing unit 126, and the scan plan generation unit 127 are connected to each other via a bus.

[0022] The input / output unit 121 has an input device and an input / output interface. The input device accepts an input operation from a user. The input / output interface inputs a signal based on the accepted input operation to the console 120. The input device is, for example, a mouse, keyboard, trackball, switch, button, joystick, touch panel, microphone, etc. The input / output interface is, for example, a data transmission interface such as optical fiber, USB, or Thunderbolt. The input / output interface may also be connected to a storage device or the like as an output device, and may read and write various data to and from the storage device. The storage device is, for example, a hard disc drive (HDD), a solid state drive (SSD), etc.

[0023] The display unit 122 has a display device and a display interface. The display device displays information to the user and a user interface for the user to input information. The user interface is, for example, a GUI (Graphical User Interface). The display interface transmits data to the display device to display images. The display device is, for example, an LCD (Liquid Crystal Display) or an organic EL (Electroluminescence) display. The display interface is, for example, a video output interface such as a DVI (Digital Visual Interface) or a High-Definition Multimedia Interface (HDMI (registered trademark), High-Definition Multimedia Interface).

[0024] The communication unit 123 connects the console 120, the bed control circuit 105, the sequence control circuit 110, and remote devices such as a server (not shown), and is capable of transmitting and receiving various data to and from each device. The communication unit 123 is configured by, for example, a wireless network adapter such as an IEEE 802.11 / Wi-Fi adapter, an adapter for communicating with a 3G, 4G / LTE, or 5G network, and a wired network adapter such as an optical fiber adapter or a power line adapter.

[0025] The storage unit 124 stores k-space data, which is data acquired in magnetic resonance imaging, reconstructed image data, and the like. The storage unit 124 also stores various parameters used in a magnetic resonance imaging assistance method, which will be described later. The storage unit 124 also stores neural network parameters. The storage unit 124 may also store various programs used by the console 120. The storage unit 124 is implemented, for example, by a storage device such as a read-only memory (ROM), a flash memory, a random access memory (RAM), a hard disk drive (HDD), a solid state drive (SSD), or a register. Flash memory, HDD, SSD, and the like are non-volatile storage media. These non-volatile storage media may also be implemented by other storage devices connected via a network, such as a network-attached storage (NAS) or an external storage server. The above-mentioned network includes, for example, the Internet, a wide area network (WAN), a local area network (LAN), a carrier network, a dedicated line, and the like.

[0026] The image reconstruction unit 125 generates image data from the scanned k-space data using a reconstruction algorithm based on Fourier transform. The image data generated by the image reconstruction unit 125 is two-dimensional slice image data, and indicates the structure of a specific slice position within the subject P. The slice position is the position of a plane on which a cross section within the subject is located. Note that the reconstruction algorithm for reconstructing the image data is not limited, and the image reconstruction unit 125 may reconstruct the image data using any reconstruction algorithm.

[0027] The image processing unit 126 performs image processing on the image data generated by the image reconstruction unit 125 to calculate information for identifying scan parameters for the actual scan. The image processing unit 126 includes an image stitching unit 61, a multiplanar reconstruction unit 62, an object detection unit 63, and a key position detection unit 64. The image stitching unit 61 generates a whole-body slice image stack in one of a sagittal plane and a coronal plane based on a plurality of partial slice image stacks (described later). The multiplanar reconstruction unit 62 reconstructs a whole-body slice image stack in an axial plane and a whole-body slice image stack in the other of a sagittal plane and a coronal plane based on the data of the whole-body slice image stack in one of the sagittal plane and the coronal plane generated by the image stitching unit 61. The object detection unit 63 detects objects in slice images included in the reconstructed whole-body slice image stack in the axial plane to detect information for generating a scan plan for the actual scan. The key position detection unit 64 detects the position of a specific part of the human body as a key position based on the whole-body slice image stacks in the sagittal plane, the coronal plane, and the axial plane. The key position is information for generating a scan plan for the actual scan.

[0028] The scan plan generation unit 127 generates a scan plan for the actual scan based on the information calculated by the image processing unit 126 and user input. The scan plan is composed of a plurality of scan parameters for performing a magnetic resonance imaging scan. Here, the scan parameters are divided into first-type scan parameters that do not need to be determined based on a scout image and can be set in advance, and second-type scan parameters that need to be determined based on a scout image.

[0029] The first type of scan parameters includes a scan sequence type, an examination range, a field of view (FOV), a stage travel distance, a matrix, a slice thickness, and a slice gap. The scan sequence type determines the display effect of various tissues in an image obtained by magnetic resonance imaging. Each type of scan sequence is applied to observe different tissues. Examples of scan sequences include T1-weighted imaging, T2-weighted imaging, fluid attenuated inversion recovery (FLAIR), and diffusion-weighted magnetic resonance imaging. The examination range is the range of a region of a subject P that needs to be examined. The examination range may be, for example, the whole body or the spine. The FOV is the size of an area that can be examined for each subscan included in a whole-body magnetic resonance scan and is the physical size of the generated image. The stage travel distance is the distance of the table travel between subscans included in a whole-body magnetic resonance scan. The matrix, slice thickness, and slice gap are used to determine the spatial resolution of an image obtained by magnetic resonance imaging.

[0030] The second type of scan parameters include the slice direction, the start and end positions of the whole-body magnetic resonance scan, the number of subscans, and the positions of the subscans. The slice direction includes coronal, sagittal, and axial, and determines the orientation of the scanned slice image. The start and end positions of the whole-body magnetic resonance scan are the start and end points of the scan in the vertical direction of the subject P. The start point of the scan is set, for example, at the vertex or the center of the head, and the end point of the scan is set, for example, at the center of the knee or the foot. The number of subscans is the number of subscans included in the whole-body magnetic resonance scan. The positions of the subscans are the positions of the scan areas of each subscan relative to the subject P.

[0031] 2 is a flowchart showing a magnetic resonance imaging assistance method according to the first embodiment. The magnetic resonance imaging assistance method of this embodiment assists a user in generating a scan plan for an actual scan based on scout images. The magnetic resonance imaging assistance method of this embodiment will be described below with reference to FIG. 2.

[0032] In steps S100 to S105, a scout scan is performed to generate a scout image of the subject P, and a whole-body slice image stack in either the sagittal plane or the coronal plane is acquired as the scout image. In the scout scan, multiple sub-scout scans are performed in sequence to cover the entire body of the subject P. The area covered by the multiple sub-scout scans is determined by setting the start and end positions of the scout scan. The start and end positions of the scout scan may be set to the top of the subject P's head and the soles of the subject P's feet, respectively, for example.

[0033] In step S100, the bed control circuit 105 moves the bed 104 on which the subject P is placed so that the top of the subject P's head is positioned in a scan region within the gradient magnetic field coil 102 having a substantially cylindrical shape.

[0034] Here, to ensure that the scout scan covers the entire body of the subject P, the subject P is positioned so that there is sufficient space between the top of his or her head and the boundary of the scan area.

[0035] When the process of step S100 is completed, the process proceeds to step S101.

[0036] In step S101, the image processing unit 126 performs a sub-scout scan to acquire partial slice images of either the sagittal plane or the coronal plane at a plurality of slice positions set in advance.

[0037] In a sub-scout scan, sequence information of a scan sequence for executing the sub-scout scan is transmitted to the sequence control circuit 110, and k-space data at a plurality of preset slice positions acquired by the sub-scout scan is received from the sequence control circuit 110. Thereafter, the image reconstruction unit 125 performs a Fourier transform on the received k-space data to acquire partial slice images at a plurality of slice positions.

[0038] The partial slice images are two-dimensional image data, and their sizes are determined by the FOV used in the scout scan. Each of the multiple partial slice images shows a structure within a local range at a respective slice position of the object P.

[0039] In this embodiment, the scout scan is either a coronal scan that generates an image of a coronal plane or a sagittal scan that generates an image of a sagittal plane. When the scout scan is a coronal scan, the plurality of pre-set slice positions are determined according to the body size and slice gap of the subject P in the anterior-posterior direction, and are positions of a plurality of parallel coronal planes, and each slice position is set at a slice gap in the anterior-posterior direction of the subject P. When the scout scan is a sagittal scan, the slice positions are determined according to the body size and slice gap of the subject P in the lateral direction, and are positions of a plurality of parallel sagittal planes, and each slice position is set at a slice gap in the lateral direction of the subject P.

[0040] Furthermore, since scout scans have low requirements for resolution and contrast, a wide FOV is adopted to shorten the scan time, and a high-speed scan sequence such as GRE (gradient echo) is selected as the scan sequence for the scout scan.

[0041] When the process of step S101 is completed, the process proceeds to step S102.

[0042] In step S102, the image processing unit 126 sequentially arranges the multiple partial slice images acquired in step S101 to generate a partial slice image stack of either a sagittal plane or a coronal plane. If the partial slice images are images of the coronal plane, the image processing unit 126 arranges the multiple partial slice images in the anterior-posterior direction of the subject P to generate a partial slice image stack. If the partial slice images are images of the sagittal plane, the image processing unit 126 arranges the multiple partial slice images in the lateral direction of the subject P to generate a partial slice image stack.

[0043] The partial slice image stack is three-dimensional image data of W×H×N, where W and H represent the width and height of the partial slice image, respectively, and N represents the number of partial slice images. The partial slice image stack shows structures within a local area at multiple slice positions of the object P.

[0044] Fig. 3 is a diagram showing an example of partial slice image stacks. (a) to (e) of Fig. 3 respectively show sagittal plane partial slice image stacks of the head and neck, chest and upper abdomen, lower abdomen and buttocks, legs, ankles and feet of a subject P. Each partial slice image stack includes sagittal plane slice images at different slice positions for a specific region of the subject P.

[0045] Returning to the description of Fig. 2, when the process of step S102 is completed, the process proceeds to step S103.

[0046] In step S103, the image processing unit 126 determines whether a predetermined number of sub-scout scans have been performed, and if it is determined that the predetermined number of sub-scout scans have not been performed, the process proceeds to step S104, and if it is determined that the predetermined number of sub-scout scans have been performed, the process proceeds to step S105. The number of sub-scout scans is determined according to the height (vertical size) of the subject P, the FOV used, and the distance the table travels. The number of sub-scout scans is set so that the range of the last sub-scout scan covers the end position of the scout scan.

[0047] In step S104, the bed control circuit 105 moves the bed 104 by a predetermined table movement distance, and moves the region of the subject P to be scanned next into the scan area of ​​the magnetic resonance imaging apparatus 100. Here, for registration between adjacent partial slice image stacks, it is preferable to set the table movement distance so that there is an overlapping portion between adjacent partial slice images in the vertical direction of the subject P. When the processing of step S104 is completed, the process proceeds to step S101.

[0048] In step S105, the image stitching unit 61 of the image processing unit 126 aligns adjacent partial slice image stacks based on the multiple partial slice images included in the partial slice image stacks, stitches the aligned partial slice image stacks together, and generates a whole-body slice image stack in either the sagittal plane or the coronal plane.

[0049] The whole-body slice image stack in one of the sagittal and coronal planes includes a plurality of sequentially arranged whole-body slice images in one of the sagittal and coronal planes, which show structures within the whole body at a plurality of slice locations within the object P.

[0050] 4 is a diagram showing an example of a whole-body slice image stack, which shows a sagittal whole-body slice image stack stitched together from partial sagittal slice image stacks of the subject P shown in FIG.

[0051] Returning to the description of Fig. 2, when the process of step S105 is completed, the process proceeds to step S106.

[0052] In step S106, the multiplanar reconstruction unit 62 of the image processing unit 126 performs multiplanar reconstruction based on the three-dimensional data included in the whole-body slice image stack of one of the sagittal and coronal planes, to generate a whole-body slice image stack of the axial plane and a whole-body slice image stack of the other of the sagittal and coronal planes. The whole-body slice image stack of the other of the sagittal and coronal planes includes a plurality of whole-body slice images of the other of the sagittal and coronal planes arranged in order. The whole-body slice image stack of the axial plane includes a plurality of whole-body slice images of the other of the axial planes arranged in order.

[0053] The following describes the process of multiplanar reconstruction using an example in which an axial plane whole-body slice image stack is generated based on a sagittal plane whole-body slice image stack. FIG. 5 is a schematic diagram showing an example in which an axial plane whole-body slice image is generated based on a sagittal plane whole-body slice image stack. As shown in FIG. 5, a plurality of reconstruction positions (positions indicated by dotted lines in the figure) spaced a certain distance apart in the vertical direction of the subject P are set in the sagittal plane whole-body slice image stack. At each reconstruction position, a whole-body axial plane slice image is generated by performing processes such as interpolation, resampling, or projection on the sagittal plane whole-body slice image at each reconstruction position. After generating the plurality of axial plane whole-body slice images, an axial plane whole-body slice image stack is generated by arranging the plurality of axial plane whole-body slice images in order in the vertical direction of the subject P.

[0054] In this embodiment, not only the whole-body slice image stack of one of the sagittal and coronal planes acquired by scout scan, but also the whole-body slice image stack of the axial plane generated by multiplanar reconstruction and the whole-body slice image stack of the other of the sagittal and coronal planes are set as scout images.

[0055] Returning to the description of Fig. 2, when the process of step S106 is completed, the process proceeds to step S107.

[0056] In steps S107 to S111, a scan plan for the actual scan is generated based on the whole-body slice image stacks of axial, sagittal, and coronal planes.

[0057] In step S107, the object detection unit 63 of the image processing unit 126 uses a trained neural network to perform object detection in each axial plane whole-body slice image included in the axial plane whole-body slice image stack. Object detection is an important technology in computer vision, and is a technology that detects whether a target is present in an image and marks the position of the target in the image with a bounding box. In this embodiment, three rough human body regions, namely, the head, torso, and lower limbs, are set as detection targets. The neural network model used for object detection is a feedforward neural network, a convolutional neural network, a Transformer, or the like.

[0058] Fig. 6 is a schematic diagram showing the results of object detection for the multiple axial whole-body slice images shown in Fig. 5. As shown in Fig. 6, in each of the axial whole-body slice images, the head, torso, or lower limbs is detected as a target, and the position of each target is displayed by a bounding box.

[0059] Returning to the description of Fig. 2, when the process of step S107 is completed, the process proceeds to step S108.

[0060] In step S108, the object detection unit 63 of the image processing unit 126 sets rough body divisions for the whole-body slice image stacks of the axial, sagittal, and coronal planes according to the type of target detected in each whole-body slice image of the axial plane. Specifically, the rough body division is set for the whole-body slice image stack of the axial plane by setting the type of detection target detected in each whole-body slice image of the axial plane to the rough body division in which the axial plane exists. Then, rough body divisions for the whole-body slice image stacks of the sagittal and coronal planes are set corresponding to the rough body division for the whole-body slice image stack of the axial plane.

[0061] 7 is a schematic diagram showing rough divisions of the human body in a sagittal slice image of the whole body set based on an axial slice image of the whole body. As shown in FIG. 7, the sagittal slice image of the whole body is divided into three regions: the head, the trunk, and the lower limbs.

[0062] Returning to the description of Fig. 2, when the process of step S108 is completed, the process proceeds to step S109.

[0063] In step S109, the key position detection unit 64 of the image processing unit 126 detects a key position of the subject P based on the whole-body slice image stacks of the axial, sagittal, and coronal planes, which are scout images. Here, the key positions are specific positions of the human body, such as the vertex, the center of the head, the spine, the center of the knees, and the soles of the feet. The key positions are used to determine second-type scan parameters, such as the start and end positions of the whole-body magnetic resonance scan in the scan plan, the number of subscans, and the positions of the subscans. In this embodiment, based on the whole-body slice image stack of the axial plane, slice images of a specific region of the subject P are selected from the whole-body slice image stacks of the sagittal and coronal planes, and segment images showing rough body sections in which the specific region is located are extracted from the selected slice images. The position of the specific region in the segment image is detected as a key position.

[0064] Below, as examples, a method for identifying the positions of the vertex and head center, a method for identifying the positions of the knee centers, and a method for identifying the position of the spine will be described.

[0065] (Identifying the location of the top and center of the head) 8 is a schematic diagram illustrating a method for determining the location of the vertex and center of the head based on a whole-body slice image stack in an axial plane and a sagittal plane. With reference to FIG. 8, a method for determining the location of the vertex and center of the head based on a whole-body slice image stack in an axial plane and a sagittal plane will be described.

[0066] First, a slice image in which the head cross-sectional area is the largest is identified from the head region of the whole-body slice image stack on the axial plane. The head cross-sectional area can be obtained from the size of the bounding box of the detected head. Note that the slice image is not limited to the slice image in which the head cross-sectional area is the largest. For example, a slice image in which the head cross-sectional area is 90% of the maximum head cross-sectional area from the head region of the whole-body slice image stack on the axial plane may be identified.

[0067] Then, one or more sagittal plane whole-body slice images are selected that are closest to the center of the head in the axial plane slice image in the left-right direction of the subject P. When selecting multiple sagittal plane whole-body slice images, slice images within a range of 2 cm from the center of the head may be selected, for example.

[0068] Then, a head image showing the head section is extracted from one selected sagittal plane whole-body slice image or an average image of multiple sagittal plane whole-body slice images, and object detection for the head is performed on the extracted head image using a trained neural network, and the head outline is enclosed in a bounding box.

[0069] Thereafter, the position of the upper frame of the bounding box of the detected head is determined as the top of the head position, and the center position of the head bounding box in the vertical direction is determined as the center position of the head.

[0070] (Identifying the knee center position) 9 is a schematic diagram showing a method for determining the location of the knee center based on a whole-body slice image stack in an axial plane and a coronal plane. A method for determining the location of the knee center based on a whole-body slice image stack in an axial plane and a coronal plane will be described with reference to FIG. 9.

[0071] First, in the upper half of the lower limb region of the whole-body slice image stack on the axial plane, the union region is obtained as the union of the regions enclosed by the respective bounding boxes for all whole-body slice images.

[0072] Thereafter, one or more coronal plane whole-body slice images located in the front part of the thigh in the acquired union region are selected in the anterior-posterior direction of the subject P.

[0073] Then, a lower limb image showing the lower limb section is extracted from one selected coronal plane whole-body slice image or an average image of multiple coronal plane whole-body slice images, and object detection for the femur is performed on the extracted lower limb image using a trained neural network, and the contour of the femur is enclosed in a bounding box.

[0074] Then, the position of the lower frame of the bounding box of the detected femur is set as the knee center position.

[0075] (locating the spine) 10 is a schematic diagram illustrating a method for determining the location of the vertebrae based on a whole-body slice image stack in axial, sagittal, and coronal planes. With reference to FIG. 10, a method for determining the location of the vertebrae based on a whole-body slice image stack in axial, coronal, and sagittal planes will be described.

[0076] First, in the middle of the trunk section of the axial plane whole-body slice image stack, the union region is obtained as the union of the regions enclosed by the respective bounding boxes for all whole-body slice images.

[0077] Thereafter, one or more sagittal whole-body slice images located in the central part of the trunk in the acquired union region are selected in the left-right direction of the subject P.

[0078] Then, a head and torso image showing the head and torso section is extracted from one selected sagittal plane whole-body slice image or an average image of multiple sagittal plane whole-body slice images, and object detection is performed on the spine in the extracted head and torso image using a trained neural network, and the outline of the spine is enclosed by a bounding box.

[0079] Then, the left frame position, right frame position, upper frame position, and lower frame position of the bounding box of the detected spine on the sagittal plane are set to the foremost, rearmost, uppermost, and lowermost ends of the spine, respectively.

[0080] Then, one or more whole-body coronal slice images located between the most distal end and the most distal end of the spine in the anterior-posterior direction of the subject P are selected in the sagittal plane.

[0081] Then, a head and torso image showing the head and torso section is extracted from one selected coronal plane whole-body slice image or an average image of multiple coronal plane whole-body slice images, and object detection for the spine is performed in the extracted head and torso image using a trained neural network, and the outline of the spine is enclosed by a bounding box.

[0082] Thereafter, the position of the left frame of the bounding box of the spine detected on the coronal plane is set as the leftmost end of the spine, and the position of the right frame of the bounding box of the spine is set as the rightmost end of the spine.

[0083] Returning to the description of Fig. 2, when the process of step S109 is completed, the process proceeds to step S110.

[0084] In step S110, the user sets the first type of scan parameters, including at least the inspection range, FOV, and stage movement distance of the actual scan, via the input device. When the process of step S110 is completed, the process proceeds to step S111.

[0085] In step S111, the scan plan generation unit 127 calculates at least one of second type scan parameters including the start position and end position of the actual scan, the number of subscans, and the positions of the subscans, based on the examination range, FOV, and table movement distance set by the user in step S110, and the key position related to the subject P identified in step S109, and generates a scan plan for the actual scan.

[0086] Specifically, first, the start and end positions of the actual scan are set based on the examination range set by the user in step S110. For example, if the user sets the examination range to the whole body, the position of the top of the head or the center of the head identified in step S109 is set as the start position of the actual scan, and the position of the center of the knee or the position of the sole of the foot identified in step S109 is set as the end position of the actual scan. Furthermore, for example, if the user sets the examination range to the spine, the positions of the uppermost and lowermost ends of the spine identified in step S109 are set as the start and end positions of the actual scan, respectively. In other words, the range to be scanned in the actual scan is set.

[0087] Then, the number of subscans required to cover the start and end positions of the actual scan is calculated from the start and end positions of the actual scan, the FOV size, and the table movement distance. Specifically, the number of subscans is calculated as the smallest m such that the sum of the FOV size and m times the table movement distance is greater than the distance between the start and end positions of the actual scan.

[0088] The scan position of the subject P in the up-down direction for each subscan is then determined so that the top end of the scan area of ​​the first subscan coincides with the start position of the actual scan, and the top end of the scan area of ​​each subsequent subscan is separated from the top end of the scan area of ​​the previous subscan by the table movement distance. The scan position of the subject P in the left-right and front-back directions for each subscan is then determined so that the central axis of each subscan in the left-right and front-back directions coincides with the central axis of the body of the subject P. The central axis of the subject P can be identified by a key position such as the spine.

[0089] Finally, a scan plan for the actual scan is generated based on the first type of scan parameters set by the user and the calculated second type of scan parameters.

[0090] When the process of step S111 is completed, the process proceeds to step S112.

[0091] In step S112, the scan plan generation unit 127 visualizes the scan plan for the actual scan and displays it on the display unit 122. Fig. 11 is a diagram showing an example of the scan plan for the actual scan displayed on the display unit 122. In Fig. 11, the start position and end position of the actual scan are indicated by dashed lines, and the position of each sub-scan is indicated by solid lines.

[0092] When the process of step S112 is completed, the process proceeds to step S113.

[0093] In step S113, the sequence control circuit 110 performs a scan on the subject P based on the scan plan for the actual scan.

[0094] When the process of step S113 is completed, the process of the magnetic resonance imaging assistance method ends.

[0095] (Effects of the embodiment) In conventional magnetic resonance imaging-assisted methods, a single slice image is used as a scout image to generate a scan plan for an actual scan, which poses a problem that the scan plan depends on the slice position of the scout image. FIG. 12 is a diagram for explaining the problem with the conventional magnetic resonance imaging-assisted method. FIG. 12 shows the start and end positions of the actual scan specified by scout images at two different slice positions. As can be seen from FIG. 12, different start and end positions of the actual scan may be specified depending on the scout images at different slice positions, which may result in inaccurate scan positions.

[0096] According to this embodiment, the start and end positions of the actual scan and the sub-scan positions are identified based on three-dimensional scout images of the axial, sagittal, and coronal planes, thereby improving the accuracy of the identified start and end positions of the actual scan and the sub-scan positions. Furthermore, according to this embodiment, appropriate slice images are selected to detect key positions, and the start and end positions of the actual scan and the sub-scan positions are identified based on the key positions, thereby further improving the accuracy of the identified start and end positions of the actual scan and the sub-scan positions.

[0097] Furthermore, according to this embodiment, a scan plan is generated using only about 1 / 20 of the slice images in the whole-body slice image stack of the axial, sagittal, and coronal planes as scout images, thereby reducing the amount of calculation required to generate the scan plan.

[0098] According to this embodiment, it is possible to assist in generating an appropriate scan plan for a production scan.

[0099] (Learning deep learning models) In the above description, the object detection unit 63 of the image processing unit 126 of this embodiment detects objects in slice images using a trained neural network. The training method of the neural network will be described below.

[0100] When neural network learning begins, first, multiple sets of pre-stored training data are read from the storage unit 124. Each set of training data includes axial, sagittal, and coronal slice images as input data, and the bounding box and type of the target as ground truth data.

[0101] Then, the multiple sets of teacher data are divided into a training set and a test set. Examples of the ratio between the training set and the test set include 80%, 20%, 90%, and 10%. For example, if the total number of sets of teacher data is 10,000, the teacher data of data #1 to #10,000 is divided into data #1 to #8,000 as the training set and data #8001 to #10,000 as the test set. Then, input data in each set of teacher data in the training set is input to a neural network, the bounding box and type of the detection target are estimated, the difference between the estimated value and the correct data is calculated, and backpropagation is performed based on the difference value to change the parameters of the neural network so that the difference between the estimated value output by the neural network and the correct answer is reduced. The above process is repeated for most of the data in the test set until the difference between the estimated value output by the neural network and the correct answer data becomes smaller than a preset threshold. After that, the neural network is determined to have completed training.

[0102] (Variation) The above has described a case where a whole-body slice image stack of an axial plane and a whole-body slice image stack of the other of the sagittal and coronal planes are generated based on a whole-body slice image stack of one of the sagittal and coronal planes, but if it is sufficient to identify a key position of the subject P based only on a whole-body slice image stack of the axial plane and a whole-body slice image stack of one of the sagittal and coronal planes, it is not necessary to generate a whole-body slice image stack of the other of the sagittal and coronal planes. According to this modification, the amount of calculation required for image processing can be reduced.

[0103] <Second embodiment> The second embodiment relates to a magnetic resonance imaging assistance method. Hereinafter, the magnetic resonance imaging assistance method according to the second embodiment will be described with reference to the drawings. In the second embodiment, differences from the first embodiment will be mainly described, and explanations of commonalities with the first embodiment will be omitted. In the description of the second embodiment, parts that are the same as those in the first embodiment will be described with the same reference numerals.

[0104] Compared with the first embodiment, the magnetic resonance imaging assistance method according to the second embodiment further includes step S114A and step S111A instead of step S111.

[0105] 13 is a flowchart showing a magnetic resonance imaging assistance method according to the second embodiment. The magnetic resonance imaging assistance method of this embodiment will be described below with reference to FIG.

[0106] In the magnetic resonance imaging assistance method of this embodiment, when the processing of step S109 is completed, the process proceeds to step S114A.

[0107] In step S114A, the object detection unit 63 of the image processing unit 126 further subdivides the torso body section into a chest section, an abdomen section, and a pelvis section, and further subdivides the lower limb body section into a leg section and an ankle section, based on the detected key positions. The chest section, abdomen section, and pelvis section may be defined, for example, by the positions of the heart and pelvis. The leg section and ankle section may be defined, for example, by the positions of the ankles. When the processing of step S114A is completed, the process proceeds to step S110.

[0108] When the process of step S110 is completed, the process proceeds to step S111A.

[0109] In step S111A, the scan plan generator 127 calculates the start position of the actual scan, the number of subscans, and the positions of the subscans based on the examination range, FOV, table movement distance, and key positions related to the subject P. It also calculates the specific absorption rate of each subscan from the subdivided body regions covered by each subscan, and generates a scan plan for the actual scan. The specific absorption rate is an important safety parameter in magnetic resonance imaging and represents the amount of RF energy absorbed by a unit mass of tissue per unit time. The magnitude of the absorption rate directly affects the risk of thermal damage that can occur in the body tissue of the subject P during a magnetic resonance imaging examination. The specific absorption rate must be set for each subdivided body region.

[0110] In the above-described embodiment, the image stitching unit 61, the multiplanar reconstruction unit 62, and the key position detection unit 64 of the image processing unit 126 are examples of an acquisition unit, a multiplanar reconstruction unit, and a first determination unit, respectively. The scan plan generation unit 127 is an example of a second determination unit. The sequence control circuit 110 is an example of an actual scan unit.

[0111] In the above-described embodiment, an example has been described in which the image stitching unit, multiplanar reconstruction unit, object detection unit, and key position detection unit in this specification are respectively realized by the image stitching unit 61, multiplanar reconstruction unit 62, object detection unit 63, and key position detection unit 64 of the image processing unit 126, but the embodiment is not limited to this. For example, the image stitching unit, multiplanar reconstruction unit, object detection unit, and key position detection unit in this specification may be realized by the image stitching unit 61, multiplanar reconstruction unit 62, object detection unit 63, and key position detection unit 64 described in the above-described embodiment, or processing units having the same functions may be realized by hardware only, software only, or a combination of hardware and software.

[0112] In the above-described embodiment, the bed control circuit 105, the sequence control circuit 110, the image processing unit 126, and the scan plan generation unit 127 are realized by a processing circuit such as a processor. In this case, the processing functions of the processing circuit are stored in the storage unit 124, for example, in the form of computer-executable programs. The processing circuit then reads and executes each program from the storage unit 124 to realize the processing function corresponding to each program. In other words, each circuit and each unit has the configuration shown in FIG. 1 when the corresponding processing circuit reads the program. Note that, although the programs corresponding to the processing functions of the processing circuit are stored in a single storage unit here, the embodiment is not limited thereto. For example, the programs corresponding to the processing functions may be stored in multiple storage units in a distributed manner, and the processing circuit may read and execute each program from each storage unit.

[0113] In the above description, the bed control circuit 105, the sequence control circuit 110, the image processing unit 126, and the scan plan generation unit 127 are each implemented by a single processing circuit, but the embodiment is not limited to this. For example, each circuit and each unit may be configured by combining multiple independent processing circuits, and each processing circuit may execute a program to implement each processing function. Furthermore, the processing functions of each circuit and each unit may be implemented by being appropriately distributed or integrated into a single or multiple processing circuits. Furthermore, the processing functions of each circuit and each unit may be implemented by a combination of hardware and software, such as circuits.

[0114] Furthermore, the term "processor" used in the above description refers to a circuit such as a central processing unit (CPU), a graphics processing unit (GPU), an application specific integrated circuit (ASIC), or a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), or a field programmable gate array (FPGA)). If the processor is a CPU, for example, the processor realizes its function by reading and executing a program stored in a memory unit. On the other hand, if the processor is an ASIC, for example, instead of storing the program in a memory unit, the function is directly incorporated into the processor circuit as a logic circuit. Note that each processor in this embodiment is not limited to being configured as a single circuit for each processor, but may be configured as a single processor by combining multiple independent circuits to realize its function. Furthermore, multiple components in FIG. 1 may be integrated into a single processor to realize its function.

[0115] Here, the program executed by the processor is provided in advance in a read-only memory (ROM) or storage unit. The program may be provided by being recorded on a computer-readable storage medium such as a compact disk (CD)-ROM, a flexible disk (FD), a recordable CD-R (CD-R), or a digital versatile disk (DVD) in a format that can be installed on these devices or in an executable format. The program may also be provided or distributed by being stored on a computer connected to a network such as the Internet and downloaded via the network. For example, the program may be composed of modules including the above-mentioned functional units. In actual hardware, a CPU reads and executes the program from a storage medium such as a ROM, whereby each module is loaded into a main memory device and generated on the main memory device.

[0116] In the above-described embodiments, the components of each device shown in the drawings are conceptual functional units and do not necessarily have to be physically configured as shown. In other words, the specific form of distribution or integration of each device is not limited to that shown in the drawings, and all or part of the devices can be functionally or physically distributed or integrated in any unit depending on various loads, usage conditions, etc. Furthermore, all or any part of the processing functions performed by each device can be realized by a CPU and a program analyzed and executed by the CPU, or can be realized as hardware using wired logic.

[0117] Furthermore, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using a known method.In addition, the information including the processing procedures, control procedures, specific names, various data and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified.

[0118] According to at least one of the embodiments described above, it is possible to assist in generating an appropriate scan plan for a production scan.

[0119] Although several embodiments have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, modifications, and combinations of embodiments can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0120] 100 Magnetic resonance imaging device 120 Console 126 Image Processing Unit 61 Image stitching section 62 Multi-section reconstruction part 63 Object detection unit 64 Key position detector

Claims

1. an acquiring step of acquiring a plurality of first slice images corresponding to different slice positions, the plurality of first slice images corresponding to one of a sagittal plane and a coronal plane, by performing a scout scan on the subject; a multiplanar reconstruction step of generating a second slice image corresponding to an axial plane by multiplanar reconstruction based on the plurality of first slice images; a first determination step of determining a position corresponding to a specific region of the subject based on the second slice image; a second determination step of determining a range to be scanned in an actual scan based on the plurality of first slice images at the positions determined in the first determination step; a main scan step of performing a scan on the subject within the range determined by the second determination step; A magnetic resonance imaging assisted method comprising:

2. the second determination step determines a start position and an end position of the scan in the actual scan.

2. The method of claim 1, wherein the method is a method for assisted magnetic resonance imaging.

3. the second determination step determines at least one of the number and positions of a plurality of sub-scans included in the main scan.

2. The method of claim 1, wherein the method is a method for assisted magnetic resonance imaging.

4. the second determining step determines at least one of the number and positions of the plurality of sub-scans based on a set size of a field of view (FOV).

4. The method of claim 3, wherein the method is a method for assisted magnetic resonance imaging.

5. The multiplanar reconstruction step generates a plurality of the second slice images, the first determining step determines a position corresponding to a specific region of the subject based on the plurality of second slice images; 2. The method of claim 1, wherein the method is a method for assisted magnetic resonance imaging.

6. the second determining step determines a range to be scanned in the actual scan based on a position of the vertex or the center of the head in a first slice image corresponding to a sagittal plane at the position determined in the first determining step.

6. A magnetic resonance imaging assistance method according to claim 1.

7. the second determining step determines a range to be scanned in the actual scan based on a position of a center of the knee in a first slice image corresponding to a coronal plane at the position determined in the first determining step.

6. A magnetic resonance imaging assistance method according to claim 1.

8. the second determining step determines a range to be scanned in the actual scan based on a position of the spine in a first slice image corresponding to at least one of a sagittal plane and a coronal plane at the position determined in the first determining step.

6. A magnetic resonance imaging assistance method according to claim 1.

9. an acquisition unit that acquires a plurality of first slice images corresponding to different slice positions, the plurality of first slice images corresponding to one of a sagittal plane and a coronal plane, by performing a scout scan on the subject; a multiplanar reconstruction unit that generates a second slice image corresponding to an axial plane by multiplanar reconstruction based on the plurality of first slice images; a first determination unit that determines a position corresponding to a specific region of the subject based on the second slice image; a second determination unit that determines a range to be scanned in an actual scan based on the plurality of first slice images at the positions determined by the first determination unit; an actual scanning unit that performs a scan on the subject within the range determined by the second determination unit; A magnetic resonance imaging apparatus comprising:

Images