Image splitting apparatus, image splitting method, and magnetic resonance imaging apparatus

Abstract

To achieve quick division of an organ in a positioning image.SOLUTION: An image division device for magnetic resonance imaging includes an acquisition unit, a provisional positioning unit, and a division unit. The acquisition unit acquires a three-dimensional positioning image or a two-dimensional positioning image of a plurality of layers of an organ. The provisional positioning unit provisionally positions a segment where the organ exists in a layer direction of a plurality of slices included in the positioning image on the basis of the positioning image. The division unit executes image division processing for the positioning image in the segment where the organ exists, and acquires a result of the division of the organ.SELECTED DRAWING: Figure 1

Description

Technical Field

[0005] ,

[0004] , ,

[0001] The embodiments disclosed in this specification and the drawings relate to an image segmentation device, an image segmentation method, and a magnetic resonance imaging device.

Background Art

[0002] Magnetic resonance examination is a kind of examination method in image diagnosis and is currently widely used in clinical practice. Magnetic Resonance Imaging (MRI) is an imaging method that rotates the atomic nuclei of a subject located in a static magnetic field by performing magnetic excitation using a high-frequency signal at the Larmor frequency, and reconstructs an image based on the nuclear magnetic resonance (NMR) signal generated along with the magnetic excitation. <00​​​​​​​​​Furthermore, for example, in conventional methods for positioning the liver region, there is a method that uses the position of the lower rib edge instead of the lower edge of the liver, but this method cannot accurately represent the position of the lower edge of the liver.

[0006] Furthermore, multi-organ abdominal scans require scan planning by referencing information on the location and size of multiple organs. For example, in the liver region, information such as the location of the diaphragm at the upper edge of the liver, and the locations of the pancreas and common bile duct are necessary. In addition, accuracy, standardization, and speed of abdominal scan planning are required to achieve efficient and even automated magnetic resonance imaging (MRI) scans. Moreover, since the patient is placed inside the MRI machine during the scan, the scan time must be kept as short as possible. Therefore, scan planning has high time requirements, demanding detection and automated planning within seconds, or even in real time. Furthermore, the scan planning requires speed in dividing and positioning the location and size of each organ.

[0007] However, conventional techniques applied to the division and positioning of the location and size of each abdominal organ require a long time to image-divide an entire organ (e.g., the entire liver). Therefore, current algorithms for dividing the location and size of each organ cannot meet the above requirements. [Prior art documents] [Patent Documents]

[0008] [Patent Document 1] U.S. Patent No. 8693760 [Patent Document 2] Chinese Patent No. 105678746 Specification [Patent Document 3] Japanese Patent Publication No. 2020-109614 [Patent Document 4] Japanese Patent Publication No. 2012-115434 [Patent Document 5] Japanese Patent Publication No. 2007-111123 [Overview of the project] [Problems that the invention aims to solve]

[0009] One of the problems that the embodiments disclosed herein and in the drawings aim to solve is to achieve rapid segmentation of organs in positioning images. However, the problems that the embodiments disclosed herein and in the drawings aim to solve are not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described later can also be positioned as other problems. [Means for solving the problem]

[0010] The image segmentation apparatus for magnetic resonance imaging according to the embodiment comprises an acquisition unit, a provisional positioning unit, and a segmentation unit. The acquisition unit acquires a three-dimensional or multi-layered two-dimensional positioning image of an organ. The provisional positioning unit provisionally positions the segment in which the organ is located in the layer direction of a plurality of slices included in the positioning image, based on the positioning image. The segmentation unit performs image segmentation processing on the positioning image within the segment in which the organ is located and acquires the segmentation result of the organ. [Brief explanation of the drawing]

[0011] [Figure 1] Figure 1 is a schematic diagram showing an example of the configuration of an image splitting apparatus for magnetic resonance imaging according to the first embodiment. [Figure 2] Figure 2 is a flowchart showing an example of the operation of an image splitting apparatus for magnetic resonance imaging according to the first embodiment. [Figure 3] Figure 3 is a schematic diagram showing an example in which the temporary positioning unit of the image division device according to the first embodiment selects a two-dimensional cross-sectional image. [Figure 4] Figure 4 is a schematic diagram showing an example of image splitting processing performed by the splitting unit of the image splitting device according to the first embodiment on a two-dimensional cross-sectional image within a segment. [Figure 5] Figure 5 is a schematic diagram showing another example in which the temporary positioning unit of the image splitting device according to the first embodiment selects a two-dimensional cross-sectional image. [Figure 6] FIG. 6 is a schematic diagram showing an example of the configuration of an image segmentation apparatus for magnetic resonance imaging according to the second embodiment. [Figure 7] FIG. 7 is a flowchart showing an example of the operation of an image segmentation apparatus for magnetic resonance imaging according to the second embodiment. [Figure 8] FIG. 8 is a schematic diagram showing an example of local features for which an optimization unit of an image segmentation apparatus according to the second embodiment performs optimization processing. [Figure 9] FIG. 9 is a schematic diagram showing another example of local features for which an optimization unit of an image segmentation apparatus according to the second embodiment performs optimization processing. [Figure 10] FIG. 10 is a schematic diagram showing an example of optimization processing performed by an optimization unit of an image segmentation apparatus according to the second embodiment. [Figure 11] FIG. 11 is a schematic diagram showing an example of the configuration of an image segmentation apparatus for magnetic resonance imaging according to the third embodiment. [Figure 12] FIG. 12 is a flowchart showing an example of the operation of an image segmentation apparatus for magnetic resonance imaging according to the third embodiment. [Figure 13] FIG. 13 is a schematic diagram showing an example of detection processing performed by a body contour detection unit of an image segmentation apparatus according to the third embodiment. [Figure 14] FIG. 14 is a schematic diagram showing Example 1 of an image segmentation apparatus for magnetic resonance imaging according to another embodiment. [Figure 15] FIG. 15 is another schematic diagram showing Example 1 of an image segmentation apparatus for magnetic resonance imaging according to another embodiment. [Figure 16] FIG. 16 is a schematic diagram showing Example 2 of an image segmentation apparatus for magnetic resonance imaging according to another embodiment. [Figure 17] FIG. 17 is another schematic diagram showing Example 2 of an image segmentation apparatus for magnetic resonance imaging according to another embodiment. [[ID=۳۷]] [Figure 18] FIG. 18 is a schematic diagram showing an example of the configuration of a magnetic resonance imaging apparatus according to the fourth embodiment.

Embodiments for Carrying Out the Invention

[0012] The embodiments described below are made to solve the above-mentioned problems, and provide an image segmentation device for magnetic resonance imaging, an image segmentation method, and a magnetic resonance imaging device. According to one embodiment, rapid segmentation of organs in a positioning image can be realized. Also, according to one embodiment, by fully utilizing the rapid segmentation, information on the complete position and size of the organ can be provided, and rapid positioning can be realized. Further, according to one embodiment, without being restricted by structural symmetry, an asymmetric structure can be detected, segmented, and positioned, and characteristic information such as the position and size of multiple organs in a three-dimensional space and accurate positioning can be provided, and automatic scan planning for abdominal multiple organs can be realized. Also, according to one embodiment, standardization and accuracy of abdominal scans, simplification of the operation flow and reduction of time can be realized, high-efficiency automatic scan planning, high-performance automatic scan, and three-dimensional rendering can be realized, and the performance of the magnetic resonance imaging device can be improved.

[0013] An image segmentation device for magnetic resonance imaging according to one embodiment includes an acquisition unit that acquires a three-dimensional or two-dimensional positioning image of a plurality of layers of an organ, and a provisional positioning unit that provisionally positions a segment in which the organ exists in the layer direction of a plurality of slices included in the positioning image based on the positioning image, and a segmentation unit that performs image segmentation processing on the positioning image within the segment in which the organ exists and acquires a segmentation result of the organ.

[0014] Also, in an image segmentation device according to one embodiment, the provisional positioning unit selects two or more cross-sectional two-dimensional images from a plurality of cross-sectional two-dimensional images acquired based on the positioning image based on a search algorithm, performs image segmentation processing on the selected cross-sectional two-dimensional images, and identifies a cross-sectional two-dimensional image corresponding to an end portion of the organ in the layer direction, thereby provisionally positioning a segment in which the organ exists in the layer direction.

[0015] Furthermore, in the image splitting device according to one embodiment, the temporary positioning unit selects the two or more cross-sectional two-dimensional images using an equally spaced selection method, a random selection method, or a selection method based on the distribution of the organs.

[0016] Furthermore, the image segmentation device according to one embodiment further includes an optimization unit that optimizes the local features of the organ based on the segmentation results of the organ.

[0017] Furthermore, in an image division device according to one embodiment, the optimization unit selects a local feature region based on the division result of the organ, and calculates and optimizes the position of the vertices of the local features of the organ by performing three-dimensional surface detection or two-dimensional edge detection based on the selected local feature region.

[0018] Furthermore, in the image segmentation device according to one embodiment, the local features of the organ are any of the six ends that define the positional range of the organ.

[0019] Furthermore, in the image splitting device according to one embodiment, the layer direction is either the head-to-foot direction, the front-to-back direction, or the left-to-right direction.

[0020] Furthermore, in the image division device according to one embodiment, the division result of the organ is data representing the outline, size, and position of the organ.

[0021] Furthermore, in the image segmentation device according to one embodiment, the search algorithm is one of the following: linear search, binary search, tree structure search, or hash search.

[0022] Furthermore, in the image splitting device according to one embodiment, the image splitting process applies an image splitting algorithm or deep learning.

[0023] Furthermore, the image division device according to one embodiment further includes a detection unit that detects the body surface area in the two-dimensional cross-sectional image based on a plurality of two-dimensional cross-sectional images acquired based on the positioning image, and the optimization unit optimizes the local features of the organ based on the body surface area and the division results of the organ.

[0024] Furthermore, in the image splitting device according to one embodiment, the organ is one of the liver, kidney, pancreas, spleen, or heart.

[0025] Furthermore, an image segmentation method for magnetic resonance imaging according to one embodiment includes the steps of: acquiring a three-dimensional or multi-layer two-dimensional positioning image of an organ; provisionally positioning the segment in which the organ is located in the layer direction of a plurality of slices included in the positioning image based on the positioning image; and performing an image segmentation process on the positioning image within the segment in which the organ is located to obtain the segmentation result of the organ.

[0026] Furthermore, a magnetic resonance imaging apparatus according to one embodiment includes the image splitting apparatus.

[0027] Furthermore, the magnetic resonance imaging apparatus according to one embodiment further includes a positioning unit that positions the organ based on the results of the organ segmentation.

[0028] Furthermore, the magnetic resonance imaging apparatus according to one embodiment further includes a planning unit that plans the position, direction, and size of the scan ROI (Region of Interest) and FOV (Field of View) based on the organ segmentation results.

[0029] Furthermore, the magnetic resonance imaging apparatus according to one embodiment further includes a rendering unit that performs three-dimensional morphological rendering of the organ based on the results of the organ segmentation.

[0030] According to one embodiment of the image segmentation device, image segmentation method, and magnetic resonance imaging apparatus for magnetic resonance imaging, rapid segmentation of organs in positioning images can be achieved. Furthermore, according to one embodiment of the image segmentation device, image segmentation method, and magnetic resonance imaging apparatus for magnetic resonance imaging, complete position and size information of organs can be provided by making full use of the rapid segmentation, thereby achieving rapid positioning. According to one embodiment of the image segmentation device, image segmentation method, and magnetic resonance imaging apparatus for magnetic resonance imaging, asymmetric structures can be detected, segmented, and positioned without being limited by structural symmetry, characteristic information such as the position and size of multiple organs in three-dimensional space and accurate positioning can be provided, and automated scanning planning of multiple abdominal organs can be realized. According to one embodiment of the image segmentation device, image segmentation method, and magnetic resonance imaging apparatus for magnetic resonance imaging, standardization and accuracy of abdominal scans, simplification and reduction of operation flow can be achieved, highly efficient automated scan planning, high-performance automated scanning and three-dimensional rendering can be achieved, and the performance of the magnetic resonance imaging apparatus can be improved.

[0031] The configuration and operation of the image splitting apparatus for magnetic resonance imaging according to this embodiment will be described below with reference to the drawings.

[0032] Furthermore, this specification and drawings will only describe and illustrate components related to the technology disclosed herein, and will omit descriptions and illustrations of other components.

[0033] Furthermore, in this specification and the drawings, components having the same or similar functions will be denoted by the same reference numeral, and redundant explanations will be omitted as appropriate.

[0034] (First embodiment) (Configuration of the image splitting device 100) Figure 1 is a schematic diagram showing an example of the configuration of an image splitting apparatus 100 for magnetic resonance imaging according to the first embodiment.

[0035] In this embodiment, the organs mainly refer to the organs of the abdomen, such as the liver, kidneys, pancreas, and spleen. However, the organs are not limited to these and may include organs such as the heart. In this embodiment, the case where the organ is the liver will be explained as an example.

[0036] As shown in Figure 1, the image splitting device 100 comprises a positioning image acquisition unit 101, a temporary positioning unit 102, and a splitting unit 110.

[0037] The positioning image acquisition unit 101 acquires three-dimensional or multi-layered two-dimensional images of organs as positioning images. For example, a series of abdominal images in three-dimensional space may be acquired by scanning and used as positioning images, and the organ (e.g., liver) region may be included within the FOV of the images. The FOV only needs to include the organ region, and may be larger than the organ region. Furthermore, the series of abdominal images in three-dimensional space may be multi-layered two-dimensional images or three-dimensional volume imaging images. Furthermore, the series of abdominal images in three-dimensional space is a series of images arranged in the layer direction. This layer direction is the direction in which each scanned slice is aligned. Furthermore, this layer direction may be, for example, the head-to-foot direction, the front-to-back direction, or the left-to-right direction of the human body.

[0038] The temporary positioning unit 102 temporarily positions the segments in the layer direction of multiple slices contained in the positioning image, based on the positioning image acquired by the positioning image acquisition unit 101, in which organs are located. For example, the temporary positioning unit 102 may, based on a search algorithm, select two or more two-dimensional cross-sectional images from among multiple two-dimensional cross-sectional images acquired based on the positioning image, perform image segmentation processing on the selected two-dimensional cross-sectional images, and identify the two-dimensional cross-sectional images corresponding to the ends of organs in the layer direction, thereby temporarily positioning the segments in the layer direction in which organs are located.

[0039] Here, if the positioning image acquired by the positioning image acquisition unit 101 is a cross-sectional scan image, the positioning image can be converted into a two-dimensional cross-sectional image. On the other hand, if the positioning image acquired by the positioning image acquisition unit 101 is not a cross-sectional scan image, multiple two-dimensional cross-sectional images are acquired by performing cross-sectional MPR (Multi-Planar Reconstruction) based on the positioning image. Known methods can be used to perform cross-sectional MPR (multi-planar reconstruction) based on the positioning image, and these will not be described in detail here.

[0040] Furthermore, the provisional positioning unit 102 may select two or more two-dimensional cross-sectional images from among multiple two-dimensional cross-sectional images acquired based on the positioning image using, for example, an equally spaced selection method, a random selection method, a selection method based on the distribution of organs, or other search algorithms.

[0041] Furthermore, the search algorithm may be ordered or unordered search algorithms such as linear search, binary search, tree structure search, or hash search, or it may be a search based on statistical rules regarding the distribution of liver locations.

[0042] The division unit 110 performs image division processing on the positioning image within the segment where the organ, which has been provisionally positioned by the provisional positioning unit 102, is located, and obtains the organ division result. Here, the image division processing can be performed by applying, for example, an image division algorithm or deep learning. The organ division result is data indicating the contour, size, and position of the organ. For example, the division unit 110 may obtain the organ division result by performing image division processing on each two-dimensional cross-sectional image from one end of the organ (e.g., the apical layer of the liver) to the other end (e.g., the basal layer of the liver) in the layer direction, that is, by performing cross-sectional organ division for each layer.

[0043] (Operation of the image splitting device 100) Next, the operation of the image splitting device 100 will be described with reference to Figures 2 to 5. Figure 2 is a flowchart showing an example of the operation of the image splitting device 100 for magnetic resonance imaging according to the first embodiment. Figure 3 is a schematic diagram showing an example of the temporary positioning unit 102 of the image splitting device 100 according to the first embodiment selecting a two-dimensional cross-sectional image. Figure 4 is a schematic diagram showing an example of the image splitting process performed by the splitting unit 110 of the image splitting device 100 according to the first embodiment on the two-dimensional cross-sectional image within a segment. Figure 5 is a schematic diagram showing another example of the temporary positioning unit 102 of the image splitting device 100 according to the first embodiment selecting a two-dimensional cross-sectional image.

[0044] As shown in Figure 2, after the image splitting device 100 starts operation, in step S100, the positioning image acquisition unit 101 acquires three-dimensional or multi-layered two-dimensional abdominal positioning images. Here, let's assume that the positioning image acquisition unit 101 acquires, for example, 31 layers of two-dimensional abdominal cross-sectional positioning images arranged in the layer direction, as shown on the left and right sides of Figure 3. Each layer shown in Figure 3 represents a single image as shown in the top, middle, and bottom of the left column in Figure 4. Also, since this example is a two-dimensional abdominal positioning image in cross-section, the layer direction corresponds to the cephalopod direction, which is the direction connecting the head and feet.

[0045] Next, in step S200, the temporary positioning unit 102 temporarily positions the segment in the layer direction in which the liver is located, based on the positioning image of the abdomen. Specifically, for example, as a first step, the temporary positioning unit 102 selects a two-dimensional positioning image of a cross-section to be used as the probe layer, based on the two-dimensional positioning images of the 31 layers acquired in step S100. As a second step, the temporary positioning unit 102 performs image segmentation processing on the two-dimensional positioning image of the cross-section to be used as the probe layer in which the liver is located. As a third step, the temporary positioning unit 102 uses a search algorithm to quickly search for the liver region based on the results of the image segmentation processing and identifies the layer in the layer direction (here, the cephaloped direction) in which the ends of the liver are located (i.e., the two-dimensional positioning image of the cross-section), thereby temporarily positioning the segment in the layer direction in which the organ is located. The temporary positioning unit 102 then repeatedly performs steps 1 to 3 while adjusting the position of the probe layer until it identifies the layer surface in which the ends of the liver (liver apex and liver base) are located.

[0046] Here, the probe layer is a representative layer, and the distribution of an organ (in this case, the liver) can be determined based on the division processing results of adjacent probe layers. For example, if, among two adjacent probe layers in the layer direction, the division processing result of one probe layer in the layer direction does not include the organ region, and the division processing result of the other probe layer in the layer direction does include the organ region, then it can be determined that one end of the organ in the layer direction is located in a layer between these two probe layers. In this case, the segment defined by these two probe layers is considered the appearing segment. Also, if, among two adjacent probe layers in the layer direction, the division processing result of one probe layer in the layer direction includes the organ region, and the division processing result of the other probe layer in the layer direction does not include the organ region, then it can be determined that the other end of the organ in the layer direction is located in a layer between these two probe layers. In this case, the segment defined by these two probe layers is considered the disappearing segment. Furthermore, if the division results of two adjacent probe layers in the layer direction do not include an organ region, it can be determined that the organ does not exist in any of the layers between these two probe layers. In this case, the segment defined by these two probe layers is considered a non-existent segment. Conversely, if the division results of two adjacent probe layers in the layer direction both include an organ region, it can be determined that the organ exists in one of the layers between these two probe layers. In this case, the segment defined by these two probe layers is considered an existing segment.

[0047] Here, the probe layer is preferably selected from the appearing and disappearing segments. By further searching the appearing and disappearing segments using a search algorithm, the apical and basal layers of the liver can be identified. There are many different methods of selection; for example, multiple two-dimensional cross-sectional images obtained from the positioning image may be randomly selected, selected at equal intervals, selected based on the distribution of organs, or selected based on other data search algorithms. The search algorithm may be any one of the following: linear search, binary search, tree structure search, or hash search.

[0048] Next, in step S300, the division unit 110 performs image division processing on the positioning image within the segment in which the liver is located and obtains the liver division result. Specifically, the division unit 110 performs liver division layer by layer (two-dimensional positioning image of the cross-section) on the segment in which the liver is located in the layer direction, that is, the segment from the top layer of the liver to the bottom layer of the liver, which was provisionally positioned in step S200, and obtains the division result.

[0049] The operation of the image segmentation device 100 of the first embodiment will be described below with reference to Figures 3 and 4. In this example, for example, with the aim of segmenting the liver, the positioning image acquisition unit 101 acquires two-dimensional abdominal positioning images of 31 layers of cross-sections arranged in the layer direction corresponding to the cephalopod direction. Here, each layer shown in Figure 3 represents a single image as shown in the top, middle, and bottom of the left column in Figure 4. In Figures 3 and 4, the dashed line layer represents the layer selected as the probe layer, and the dotted line layer represents the layer in which the image segmentation result will include an organ.

[0050] In this application example, as shown in Figure 3, the positioning image acquisition unit 101 acquires a two-dimensional abdominal positioning image of 31 layers arranged in the layer direction. Then, the provisional positioning unit 102 selects, for example, the layer closest to the head and the layer closest to the feet as probe layers, and further selects six probe layers at equal intervals (the dashed lines shown on the left side of Figure 3), and performs image segmentation processing on each probe layer. That is, the provisional positioning unit 102 performs image segmentation processing on the two-dimensional cross-sectional image as shown on the left side of Figure 4. Here, the probe layers that segment the liver region are represented by the dashed-dotted lines. As a result of this image segmentation processing, for example, the segmentation results of four out of eight probe layers will include the liver region (the dashed-dotted lines shown on the right side of Figure 3). Thus, as shown on the right side of Figure 3, from top to bottom in the layer direction, the non-existent segment, appearing segment, present segment, present segment, present segment, disappearing segment, and non-existent segment are defined in order. Next, the temporary positioning unit 102 further places probe layers for the appearing and disappearing segments and searches using a search algorithm to identify the liver apex layer (the first dotted-dash layer from the top in Figure 4) and the liver base layer (the first dotted-dash layer from the bottom in Figure 4). Then, the division unit 110 divides each layer surface within the segment defined by the liver apex layer and liver base layer, and obtains information such as the outline, position, and size of the liver in three-dimensional space (see the figures in the middle column in Figure 4) as the liver division result. Note that in the example shown in Figure 3, the layer closest to the head and the layer closest to the feet do not necessarily have to be selected as probe layers.

[0051] As described above, according to the image division device 100 for magnetic resonance imaging according to the first embodiment, the positioning image acquisition unit 101 acquires three-dimensional or multi-layered two-dimensional positioning images of the organ. The provisional positioning unit 102 provisionally positions the segment in the layer direction in which the organ exists based on the positioning images. The division unit 110 then performs image division processing on each positioning image within the segment in which the organ exists and obtains the organ division result. In this way, the provisional positioning unit 102 performs processing to provisionally position the segment in the layer direction in which the organ exists with a small amount of image division processing, and then performs image division processing on each cross-sectional two-dimensional image within the segment in which the organ exists. This makes it possible to quickly provisionally position the organ with a small amount of computation, reduce the range in which image division processing is required, reduce the amount of computation required for organ division, improve processing speed, and shorten processing time. On the other hand, by performing image division processing on each cross-sectional two-dimensional image within the segment in which the organ exists and obtaining the organ division result, it is possible to provide more complete spatial information such as the position and size of the organ. In other words, according to the image segmentation device 100 of the first embodiment, more complete spatial information such as the position and size of organs can be provided quickly, enabling rapid positioning. Furthermore, asymmetric structures can be detected, segmented, and positioned without being limited by structural symmetry, providing characteristic information such as the position and size of multiple organs in three-dimensional space and accurate positioning, thereby improving the performance of the magnetic resonance imaging device.

[0052] Furthermore, while Figure 3 shows an example where probe layers are selected at equal intervals, and the segment containing the liver is identified using the image segmentation results and the search algorithm, this embodiment is not limited to this. For example, as shown in Figure 5, instead of uniformly selecting probe layers at equal intervals, the distribution of organs in the layer direction may be considered, and more probe layers may be selected for segments where the ends of organs are likely to be present, and fewer probe layers may be selected for segments where organs are likely to appear continuously and segments where organs are unlikely to be present. In the example shown in Figure 5, the layer closest to the head and the layer closest to the feet do not necessarily have to be selected as probe layers. This allows for rapid provisional positioning of organs with less computation, further reduces the range requiring image segmentation, and further decreases the amount of computation required for organ segmentation. As a result, processing speed can be further improved and processing time can be further reduced.

[0053] (Second embodiment) (Configuration of the image splitting device 100a) Figure 6 is a schematic diagram showing an example of the configuration of an image splitting device 100a for magnetic resonance imaging according to the second embodiment. In this embodiment, as with the first embodiment, the case where the organ is the liver will be used as an example. In this embodiment, the same reference numerals will be used for components that are the same as or similar to those in the first embodiment, and detailed descriptions will be omitted, with only the differences being described in detail.

[0054] As shown in Figure 6, the image splitting device 100a includes a positioning image acquisition unit 101, a temporary positioning unit 102, an optimization unit 103, and a splitting unit 110.

[0055] Here, the positioning image acquisition unit 101, the temporary positioning unit 102, and the division unit 110 are the same as in the first embodiment, so their explanation will be omitted.

[0056] The optimization unit 103 optimizes the local features of the organ based on the organ division results obtained by the division unit 110. Here, the local features of the organ are, for example, one of the six ends that define the position range of the organ. For example, the optimization unit 103 may select a local feature region based on the organ division results and calculate and optimize the position of the vertices of the organ's local features by performing three-dimensional surface detection or two-dimensional edge detection based on the selected local feature region. The vertices referred to here may be, for example, the center, centroid, or other boundary points obtained by calculation methods such as weighted average of the local feature region.

[0057] Furthermore, taking the liver as an example, because the liver is close to the diaphragm, it is affected by the movement of the diaphragm. Therefore, when it is necessary to consider the effect of respiration on the position of the liver apex, the requirements for edges and boundary points become higher, and it is necessary to detect the liver apex region more accurately and pinpoint the position of the apex in that liver apex region with precision. Also, when it is necessary to determine the morphological size of the liver, for example, in order to determine liver enlargement, it is necessary to measure the distance between the liver apex and the liver base. In this case, it is necessary to perform more detailed detection of the liver base as well and pinpoint the position of the apex in the liver base region with precision. The optimization unit 103 is preferably applied in such cases.

[0058] (Operation of the image splitting device 100a) Next, the operation of the image splitting device 100a will be described with reference to Figures 7 to 10. Figure 7 is a flowchart showing an example of the operation of the image splitting device 100a for magnetic resonance imaging according to the second embodiment. Figure 8 is a schematic diagram showing an example of a local feature that the optimization unit 103 of the image splitting device 100a according to the second embodiment performs optimization processing on. Figure 9 is a schematic diagram showing another example of a local feature that the optimization unit 103 of the image splitting device 100a according to the second embodiment performs optimization processing on. Figure 10 is a schematic diagram showing an example of the optimization processing performed by the optimization unit 103 of the image splitting device 100a according to the second embodiment.

[0059] Here, steps S100 to S300 shown in Figure 7 are the same as steps S100 to S300 shown in Figure 2, so their explanation will be omitted.

[0060] As shown in Figure 7, in step S300, the image splitting device 100a has a splitting unit 110 that, based on the segments in the layer direction in which the liver is located, which were provisionally positioned in step S200, i.e., segments from the top layer of the liver to the bottom layer of the liver, splits the liver layer by layer for each segment and obtains the splitting result. However, without ending the process, it proceeds to step S400.

[0061] Next, in step S400, the optimization unit 103 optimizes the local characteristics of the liver based on the liver segmentation results obtained in step S300.

[0062] Here, the local features of the liver are, for example, one of the six ends that define the location range of the liver, as shown in Figures 8 and 9: the anterior end A, the posterior end P, the right end R, the left end L, the apical end (also called the liver apical boundary point) H, and the basal end (also called the liver base boundary point) F.

[0063] Specifically, the optimization unit 103 first extracts the liver apex boundary point H based on the liver segmentation results. Then, using the extracted liver apex boundary point H as a reference point, the optimization unit 103 selects a fine region of the liver apex as a local feature region based on the size and location of the patient's liver region and the size and location of the body contour. Next, the optimization unit 103 reconstructs the liver apex surface, which will be the liver apex layer, by performing three-dimensional surface detection or two-dimensional edge detection within the selected local feature region, for example, as shown in Figure 10. Next, the optimization unit 103 selects a liver apex surface or a set of liver apex surfaces composed of several adjacent layers based on a threshold set according to the clinical accuracy requirements. Finally, the optimization unit 103 completes the optimization by comprehensively calculating the central position of the liver apex based on the selected liver apex surface or set of liver apex surfaces. Here, as a specific calculation method for calculating the central position of the liver apex, for example, a method for calculating the center of multiple points, centroid, or weighted average can be used.

[0064] Although the liver apex boundary point H was used as an example in this explanation, the optimization unit 103 may apply a similar method to other ends of the liver to perform optimization.

[0065] As described above, the image segmentation device 100a for magnetic resonance imaging according to the second embodiment includes, in addition to the configuration of the first embodiment, an optimization unit 103 that more precisely detects and optimizes local features of organs based on the organ segmentation results. This provides the same effects as the first embodiment, as well as more accurate and detailed characteristic information such as the position and size of organs in three-dimensional space, and more accurate positioning of organs in three-dimensional space, thereby further improving the performance of the magnetic resonance imaging device.

[0066] (Third embodiment) (Configuration of image splitting device 100b) Figure 11 is a schematic diagram showing an example of the configuration of an image splitting device 100b for magnetic resonance imaging according to the third embodiment. In this embodiment, as with the first and second embodiments, the case where the organ is the liver will be used as an example. In this embodiment, the same reference numerals are used for components that are the same as or similar to those in the first and second embodiments, and detailed descriptions are omitted, with only the differences being described in detail.

[0067] As shown in Figure 11, the image splitting device 100b comprises a positioning image acquisition unit 101, a temporary positioning unit 102, an optimization unit 103, a body contour detection unit 104, and a splitting unit 110.

[0068] Here, the positioning image acquisition unit 101, the temporary positioning unit 102, and the division unit 110 are the same as in the first and second embodiments, so their explanation will be omitted.

[0069] The body contour detection unit 104 detects the body surface area in the two-dimensional cross-sectional image based on a plurality of two-dimensional cross-sectional images acquired based on the positioning image acquired by the positioning image acquisition unit 101.

[0070] In this embodiment, the optimization unit 103 optimizes the local features of the organ based on the body surface range detected by the body contour detection unit 104 and the organ division results obtained by the division unit 110.

[0071] (Operation of the image splitting device 100b) Next, the operation of the image splitting device 100b will be described with reference to Figures 12 and 13. Figure 12 is a flowchart showing an example of the operation of the image splitting device 100b for magnetic resonance imaging according to the third embodiment. Figure 13 is a schematic diagram showing an example of the detection process performed by the body contour detection unit 104 of the image splitting device 100b according to the third embodiment.

[0072] Here, steps S100 to S300 shown in Figure 12 are the same as steps S100 to S300 shown in Figure 2, so their explanation is omitted.

[0073] As shown in Figure 12, in step S500, the image segmentation device 100b uses a body contour detection unit 104 to detect the body surface area in a two-dimensional cross-sectional image based on multiple two-dimensional cross-sectional images obtained from the positioning image acquired in step S100. For example, as shown by the thick-lined frame in Figure 13, the frame touching the body surface in the two-dimensional cross-sectional image is detected as the body surface area. Here, various methods such as known edge detection methods or grayscale value analysis methods can be applied to detect the body surface area in a two-dimensional cross-sectional image, and these will not be described in detail here.

[0074] In step S400, the optimization unit 103 optimizes the local characteristics of the liver based on the body surface range detected in step S500 and the liver segmentation results obtained in step S300.

[0075] As described above, the image segmentation device 100b for magnetic resonance imaging according to the third embodiment includes, in addition to the configuration of the first and second embodiments, a body contour detection unit 104 for detecting body surface contours in a two-dimensional cross-sectional image. This allows for the removal of arm areas, rapid and efficient positioning of the anterior-posterior-lateral (APLR) centers and anatomically valid structural areas, and the removal of data that interferes with the optimization unit 103. As a result, in addition to the same effects as the first and second embodiments, it is possible to provide more accurate and detailed characteristic information such as the position and size of organs in three-dimensional space, and to further provide more accurate positioning of organs in three-dimensional space, thereby further improving the performance of the magnetic resonance imaging device.

[0076] Although several embodiments have been described above using the liver as an example, the embodiments of the technology disclosed in this application are not limited to these and may be realized by other embodiments. Examples 1 and 2 of an image splitting device according to other embodiments are described below.

[0077] Here, in Examples 1 and 2, the flowcharts of the configuration and operation described in any of the first to third embodiments can be used. Below, the flowchart of the configuration and operation of the first embodiment will be used for explanation.

[0078] (Example 1) First, Example 1 will be described with reference to Figures 14 and 15. Figure 14 is a schematic diagram showing Example 1 of an image splitting apparatus for magnetic resonance imaging according to another embodiment. Figure 15 is another schematic diagram showing Example 1 of an image splitting apparatus for magnetic resonance imaging according to another embodiment.

[0079] In Example 1, as shown in Figure 14, the positioning image acquisition unit 101 acquires three-dimensional or multi-layered two-dimensional positioning images of the abdomen of an organ. Here, the positioning image acquisition unit 101 acquires, for example, two-dimensional abdominal positioning images of multiple layers of cross-sections arranged in the layer direction, as shown on the left and right sides of Figure 14. Each layer shown in Figure 14 represents a single image as shown on the left side of Figure 15. Here, since these are two-dimensional abdominal positioning images of cross-sections, the layer direction corresponds to the cephalopedial direction, which is the direction connecting the head and feet. In Example 1, the division of a kidney is used as an example.

[0080] In detail, in Example 1, the positioning image acquisition unit 101 acquires a two-dimensional abdominal positioning image of multiple layers arranged in the layer direction. Here, the kidneys are often distributed in the lower part of the abdomen (for example, in the lower 1 / 2 to 1 / 3 of the way up). Therefore, the provisional positioning unit 102 selects probe layers densely in the lower 1 / 2 to 1 / 3 of the two-dimensional abdominal positioning image of multiple layers arranged in the layer direction, and probe layers sparsely in the other areas (dashed lines shown on the left side of Figure 14), and performs image segmentation processing on each probe layer. That is, the provisional positioning unit 102 performs image segmentation processing on the two-dimensional cross-sectional image as shown on the left side of Figure 15. Here, the probe layers that have segmented the kidney region are represented by dashed lines. As a result of this image segmentation processing, for example, the segmentation results of 3 out of 8 probe layers will include the kidney region (dashed lines shown on the right side of Figure 14). As a result, as shown on the right side of Figure 14, the following segments are defined sequentially from top to bottom in the layer direction: non-existent segment, non-existent segment, appearing segment, presenting segment, presenting segment, disappearing segment, and non-existent segment. Note that in the example shown in Figure 14, the layer closest to the head and the layer closest to the feet do not necessarily have to be selected as probe layers. Next, the provisional positioning unit 102 further sets probe layers for the appearing and disappearing segments and searches using a search algorithm to identify the apical layer of the kidney (the first dotted-dash layer from the top in Figure 14) and the basal layer of the kidney (the first dotted-dash layer from the bottom in Figure 14). Then, the division unit 110 divides each layer surface within the segment defined by the apical and basal layers of the kidney (see the dotted-dash layers shown in Figure 15), and obtains information such as the outline, position, and size of the kidney in three-dimensional space (see the diagram shown in the center of Figure 15) as the result of the kidney division.

[0081] Thus, according to Example 1, in addition to the same effects as in the first embodiment, the probe layer can be selected more accurately based on the rules in the statistics of kidney position distribution. This allows for rapid provisional positioning of organs with less computation, further reduces the area requiring image segmentation processing, further decreases the amount of computation required for organ segmentation, further improves processing speed, and further shortens processing time.

[0082] (Example 2) Next, Example 2 will be described with reference to Figures 16 and 17. Figure 16 is a schematic diagram showing Example 2 of an image splitting apparatus for magnetic resonance imaging according to another embodiment. Figure 17 is another schematic diagram showing Example 2 of an image splitting apparatus for magnetic resonance imaging according to another embodiment.

[0083] In Example 2, as shown in Figure 15, the positioning image acquisition unit 101 acquires three-dimensional or multi-layered two-dimensional abdominal positioning images of the organs. Here, the positioning image acquisition unit 101 acquires, for example, multi-layered coronal plane two-dimensional abdominal positioning images arranged in the layer direction, as shown on the left and right sides of Figure 16. Each layer shown in Figure 16 represents a single image as shown on the left side of Figure 17. Here, since these are two-dimensional coronal plane abdominal positioning images, the layer direction corresponds to the anterior-posterior direction, which connects the front and back of the body. In Example 2, the division of the kidney is used as an example.

[0084] In detail, in Example 2, the positioning image acquisition unit 101 acquires a two-dimensional positioning image of the abdomen in the coronal plane of multiple layers arranged in the layer direction. Here, the kidneys are often distributed in the posterior part of the abdomen (for example, the posterior 2 / 5 of the way from the back). Therefore, the provisional positioning unit 102 selects probe layers densely for the posterior 2 / 5 of the two-dimensional positioning image of the abdomen in the coronal plane of multiple layers arranged in the layer direction, and probe layers sparsely for other areas (the dashed lines shown on the left side of Figure 16), and performs image segmentation processing on the probe layers. That is, the provisional positioning unit 102 performs image segmentation processing on the two-dimensional image of the coronal plane as shown on the left side of Figure 17. Here, the probe layers that have segmented the kidney region are represented by the dashed-dotted lines. As a result of this image segmentation processing, for example, the segmentation results of two out of six probe layers will include the kidney region (the dashed-dotted lines shown on the right side of Figure 16). As a result, as shown on the right side of Figure 16, the non-existent segment, appearing segment, present segment, disappearing segment, and non-existent segment are defined sequentially from front to back in the layer direction. Note that in the example shown in Figure 16, the layer closest to the front and the layer closest to the back do not necessarily have to be selected as probe layers. Next, the provisional positioning unit 102 further sets probe layers for the appearing and disappearing segments and searches using a search algorithm to identify the pre-renal layer (the dotted-dash layer from the left in Figure 17) and the posterior renal layer (the dotted-dash layer from the right in Figure 17). Then, the division unit 110 divides each layer surface within the ligment defined by the pre-renal and posterior renal layers (see the dotted-dash layers in Figure 17) and obtains information such as the outline, position, and size of the liver in three-dimensional space (see the figure in the center of Figure 17) as the result of the kidney division.

[0085] Thus, according to Example 2, in addition to the same effects as in the First Embodiment and Example 1, the kidney can be rapidly divided and positioned using coronal plane positioning.

[0086] Furthermore, although the embodiments and examples described above illustrate cases in which the technology disclosed by this application is realized by an image splitting device and an image splitting method, the technology disclosed by this application may also be realized by a magnetic resonance imaging device.

[0087] For example, the technology disclosed herein can be realized as a magnetic resonance imaging apparatus equipped with an image splitting device described in any of the embodiments and examples described above. Hereinafter, an example of applying the technology disclosed herein to a magnetic resonance imaging apparatus will be described as a fourth embodiment.

[0088] (Fourth embodiment) (Configuration of Magnetic Resonance Imaging System 200) Figure 18 is a schematic diagram showing an example of the configuration of a magnetic resonance imaging apparatus according to the fourth embodiment.

[0089] As shown in Figure 18, the magnetic resonance imaging apparatus 200 comprises a static magnetic field magnet 1, gradient magnetic field coils 2, gradient magnetic field power supply 3, whole-body RF coil 4, local RF coil 5, transmitting circuit 6, receiving circuit 7, RF shield 8, stand 9, patient table 10, input interface 11, display 12, memory circuit 13, image splitting device 100, and processing circuits 14-17.

[0090] The static magnetic field magnet 1 generates a static magnetic field in the space within the bore 9a of the base 9. Specifically, the static magnetic field magnet 1 is formed in a hollow, substantially cylindrical shape (including those with an elliptical cross-sectional shape perpendicular to the central axis), and generates a static magnetic field in the space within the bore 9a located on its inner circumference. For example, the static magnetic field magnet 1 is a superconducting magnet or a permanent magnet. The superconducting magnet referred to here is composed of, for example, a container filled with a coolant such as liquid helium and a superconducting coil immersed in the container.

[0091] The gradient magnetic field coil 2 is positioned inside the static magnetic field magnet 1 and generates a gradient magnetic field within the bore 9a of the frame 9. Specifically, the gradient magnetic field coil 2 is formed in a hollow, substantially cylindrical shape (including those with an elliptical cross-sectional shape perpendicular to the central axis) and has X-coils, Y-coils, and Z-coils corresponding to the mutually orthogonal X-axis, Y-coils, and Z-coils, respectively. The X-coils, Y-coils, and Z-coils generate a gradient magnetic field that changes linearly along each axial direction based on the current supplied from the gradient magnetic field power supply 3. Here, the Z-axis is set to align with the magnetic flux of the static magnetic field generated by the static magnetic field magnet 1. The X-axis is set to align with the horizontal direction perpendicular to the Z-axis, and the Y-axis is set to align with the vertical direction perpendicular to the Z-axis. Here, the X-axis, Y-axis, and Z-axis constitute the device coordinate system unique to the magnetic resonance imaging apparatus 200.

[0092] The gradient power supply 3 generates the aforementioned gradient magnetic field by supplying current to the gradient coil 2. Specifically, the gradient power supply 3 supplies current individually to the X coil, Y coil, and Z coil of the gradient coil 2, thereby generating a gradient magnetic field that changes linearly along the mutually orthogonal readout direction, phase encoding direction, and slice direction, respectively. Here, the axis along the readout direction, the axis along the phase encoding direction, and the axis along the slice direction constitute a logical coordinate system for defining the slice region or volume region to be imaged.

[0093] Specifically, the gradient magnetic fields along the readout direction, phase encoding direction, and slice direction are superimposed on the static magnetic field generated by the static magnetic field magnet 1, thereby adding spatial positional information to the NMR signal generated from the subject S. Specifically, the gradient magnetic field in the readout direction adds positional information in the readout direction to the NMR signal by changing the frequency of the NMR signal according to the position in the readout direction. The gradient magnetic field in the phase encoding direction adds positional information in the phase encoding direction to the NMR signal by changing the phase of the NMR signal according to the position in the phase encoding direction. The gradient magnetic field in the slice direction adds positional information in the slice direction to the NMR signal. For example, when imaging of slice cross-sections (2D imaging) is performed, the gradient magnetic field in the slice direction is used to determine the direction, thickness, and number of slice cross-sections, and when imaging volume data (3D imaging) is performed, it is used to change the phase of the NMR signal according to the position in the slice direction.

[0094] The whole-body RF coil 4 is positioned on the inner circumference side of the gradient coil 2 and applies RF pulses (excitation pulses, etc.) to the subject S placed in the bore 9a of the frame 9, and receives the NMR signal (echo signal, etc.) generated from the subject S due to the influence of the RF pulses. Specifically, the whole-body RF coil 4 is formed in a hollow, approximately cylindrical shape (including those with an elliptical cross-section perpendicular to the central axis), and applies RF pulses to the subject S placed in the bore 9a of the frame 9 based on the RF pulse signal supplied from the transmitting circuit 6. The whole-body RF coil 4 then receives the NMR signal generated from the subject S due to the influence of the RF pulses and outputs the received NMR signal to the receiving circuit 7. For example, the whole-body RF coil 4 is a birdcage type coil or a TEM (Transverse Electromagnetic) coil.

[0095] The local RF coil 5 is placed in the bore 9a of the rig 9 together with the subject S during imaging and receives the NMR signal generated from the subject S. Specifically, a local RF coil 5 is prepared for each part of the subject S and is placed near the area to be imaged when imaging of the subject S is performed, and receives the NMR signal generated from the subject S due to the influence of the RF pulse applied by the whole-body RF coil 4. The local RF coil 5 then converts the received NMR signal from an analog signal to a digital signal to generate NMR data and outputs the generated NMR data to the processing circuit 15. For example, the local RF coil 5 is a surface coil or a phased array coil configured by combining multiple surface coils as coil elements.

[0096] The transmitting circuit 6 outputs an RF pulse signal corresponding to the resonance frequency (Larmor frequency) unique to the target atomic nucleus placed in a static magnetic field to the whole-body RF coil 4. Specifically, the transmitting circuit 6 includes a pulse generator, an RF generator, a modulator, and an amplifier. The pulse generator generates the waveform of the RF pulse signal. The RF generator generates an RF signal at the resonance frequency. The modulator generates an RF pulse signal by modulating the amplitude of the RF signal generated by the RF generator with the waveform generated by the pulse generator. The amplifier amplifies the RF pulse signal generated by the modulator and outputs it to the whole-body RF coil 4.

[0097] The receiving circuit 7 generates NMR data based on the NMR signal output from the whole-body RF coil 4 and outputs the generated NMR data to the processing circuit 15. Specifically, the receiving circuit 7 includes a selector, a preamplifier, a phase detector, and an A / D converter. The selector selectively inputs the NMR signal output from the whole-body RF coil 4. The preamplifier amplifies the NMR signal output from the selector. The phase detector detects the phase of the NMR signal output from the preamplifier. The A / D converter converts the analog signal output from the phase detector into a digital signal to generate NMR data and outputs the generated NMR data to the processing circuit 15. Note that, as described here, the processing performed by the receiving circuit 7 does not necessarily have to be performed entirely by the receiving circuit 7; some processing (for example, processing by the A / D converter) may be performed by the whole-body RF coil 4.

[0098] The RF shield 8 is positioned between the gradient magnetic field coil 2 and the whole-body RF coil 4, shielding the gradient magnetic field coil 2 from RF pulses generated by the whole-body RF coil 4. Specifically, the RF shield 8 is formed in a hollow, substantially cylindrical shape (including those with an elliptical cross-section perpendicular to the central axis of the cylinder), and is positioned in the space on the inner circumference side of the gradient magnetic field coil 2 so as to cover the outer surface of the whole-body RF coil 4.

[0099] The rig 9 has a hollow bore 9a formed in a substantially cylindrical shape (including those with an elliptical cross-sectional shape perpendicular to the central axis), and houses a static magnetic field magnet 1, a gradient magnetic field coil 2, a whole-body RF coil 4, and an RF shield 8. Specifically, the rig 9 houses the whole-body RF coil 4 on the outer circumference of the bore 9a, the RF shield 8 on the outer circumference of the whole-body RF coil 4, the gradient magnetic field coil 2 on the outer circumference of the RF shield 8, and the static magnetic field magnet 1 on the outer circumference of the gradient magnetic field coil 2. Here, the space within the bore 9a of the rig 9 becomes the imaging area where the subject S is placed during imaging.

[0100] The examination bed 10 is equipped with a top plate 10a on which the subject S is placed, and when imaging of the subject S is performed, the top plate 10a on which the subject S is placed is moved into the bore 9a of the frame 9. For example, the examination bed 10 is installed so that the longitudinal direction of the top plate 10a is parallel to the central axis of the static magnetic field magnet 1.

[0101] Here, we describe an example in which the magnetic resonance imaging apparatus 200 has a so-called tunnel-type structure in which the static magnetic field magnet 1, gradient magnetic field coil 2, and whole-body RF coil 4 are each formed in a substantially cylindrical shape, but the embodiments are not limited to this. For example, the magnetic resonance imaging apparatus 200 may have a so-called open-type structure in which a pair of static magnetic field magnets, a pair of gradient magnetic field coils, and a pair of RF coils are arranged facing each other across the imaging area in which the subject S is placed. In such an open-type structure, the space sandwiched between the pair of static magnetic field magnets, the pair of gradient magnetic field coils, and the pair of RF coils corresponds to the bore in the tunnel-type structure.

[0102] The input interface 11 receives various instructions and information input operations from the operator. Specifically, the input interface 11 is connected to the processing circuit 17 and converts the input operations received from the operator into electrical signals and outputs them to the processing circuit 17. For example, the input interface 11 can be implemented by a trackball for setting imaging conditions and scan ROI, a switch button, a mouse, a keyboard, a touchpad for input operations by touching the operating surface, a touchscreen that integrates a display screen and a touchpad, a non-contact input circuit using an optical sensor, and an audio input circuit. In this specification, the input interface 11 is not limited to those equipped with physical operating components such as a mouse or keyboard. For example, an electrical signal processing circuit that receives electrical signals corresponding to input operations from an external input device provided separately from the device and outputs these electrical signals to a control circuit is also included as an example of the input interface 11.

[0103] The display 12 displays various types of information. Specifically, the display 12 is connected to the processing circuit 17 and converts the data of various types of information sent from the processing circuit 17 into electrical signals for display and outputs them. For example, the display 12 can be implemented as an LCD monitor, a CRT monitor, a touch panel, or the like.

[0104] The memory circuit 13 stores various types of data. Specifically, the memory circuit 13 is connected to processing circuits 14-17 and stores various types of data input and output by each processing circuit. For example, the memory circuit 13 can be implemented using semiconductor memory elements such as RAM (Random Access Memory) or flash memory, or a hard disk or optical disc.

[0105] The processing circuit 14 has a bed control function 14a. The bed control function 14a controls the operation of the bed 10 by outputting control electrical signals to the bed 10. For example, the bed control function 14a receives instructions from the operator via the input interface 11 to move the top plate 10a in the longitudinal direction, vertical direction, or horizontal direction, and operates the bed 10's top plate 10a movement mechanism to move the top plate 10a according to the received instructions.

[0106] The processing circuit 15 has an acquisition function 15a. The acquisition function 15a acquires k-space data of the subject S based on the imaging sequence output from the processing circuit 17. Specifically, the acquisition function 15a acquires NMR data by driving the gradient power supply 3, the transmitting circuit 6, the receiving circuit 7, and the local RF coil 5 according to various imaging sequences output from the processing circuit 17. Here, the imaging sequence is information that defines the timing and strength of the current supplied by the gradient power supply 3 to the gradient coil 2, the timing and strength of the RF pulse signal supplied by the transmitting circuit 6 to the whole-body RF coil 4, and the timing of the NMR signal sampling by the receiving circuit 7. The acquisition function 15a then stores the NMR data output from the receiving circuit 7 and the local RF coil 5 in the storage circuit 13. Here, the NMR data stored in the storage circuit 13 is stored as k-space data representing a two-dimensional or three-dimensional k-space, with positional information along the readout direction, phase encoding direction, and slice direction assigned by each of the aforementioned gradient magnetic fields.

[0107] The processing circuit 16 has a generation function 16a. The generation function 16a generates an image from the k-space data of the subject S collected by the collection function 15a of the processing circuit 15. Specifically, the generation function 16a reads the k-space data collected by the collection function 15a of the processing circuit 15 from the storage circuit 13, and generates a two-dimensional or three-dimensional image by applying reconstruction processing such as a Fourier transform to the read k-space data. Then, the generation function 16a stores the generated image in the storage circuit 13.

[0108] The processing circuit 17 receives imaging conditions from the operator via the input interface 11 and generates an imaging sequence for collecting k-space data of the subject S based on the input imaging conditions. The processing circuit 17 also outputs the generated imaging sequence to the processing circuit 15, causing the acquisition function 15a of the processing circuit 15 to collect k-space data. The processing circuit 17 also controls the processing circuit 16 to reconstruct an image from the k-space data collected by the acquisition function 15a of the processing circuit 15. In response to a request from the operator, the processing circuit 17 reads an image stored in the memory circuit 13 and displays the read image on the display 12.

[0109] Here, for example, processing circuits 14 to 17 are each implemented by a processor. In this case, the processing functions of each processing circuit are stored in the memory circuit 13 in the form of a program that can be executed by a computer. Then, each processing circuit implements the processing function corresponding to each program by reading and executing each program from the memory circuit 13. In other words, each processing circuit, in the state where each program has been read, will have the processing functions shown in each processing circuit of Figure 18.

[0110] Furthermore, the processing circuits 14 to 17 are not limited to being implemented by a single processor. For example, each processing circuit may be composed of a combination of multiple independent processors, with each processor executing a program to realize each processing function. Also, the processing functions of each processing circuit may be implemented by being appropriately distributed or integrated across one or more processing circuits. In addition, although the above description assumed that a single memory circuit 13 stores the program corresponding to each processing function, the embodiments are not limited to this. For example, multiple memory circuits may be distributed and arranged for each processing circuit, and each processing circuit may read the corresponding program from its individual memory circuit.

[0111] Under this configuration, the magnetic resonance imaging apparatus 200 according to this embodiment includes the image splitting apparatus 100 described in the first embodiment described above. Here, an example is given in which the image splitting apparatus 100 described in the first embodiment is used, but this embodiment is not limited to this, and the magnetic resonance imaging apparatus 200 may also include the image splitting apparatus 100a described in the second embodiment, or the image splitting apparatus 100b described in the third embodiment.

[0112] Furthermore, the magnetic resonance imaging apparatus 200 according to this embodiment includes a processing circuit 17 comprising a positioning function 17a, a planning function 17b, and a rendering function 17c. Here, the positioning function 17a is an example of a positioning unit. The planning function 17b is an example of a planning unit. The rendering function 17c is an example of a rendering unit.

[0113] The positioning function 17a positions the organ based on the organ division result acquired by the division unit 110 of the image division device 100.

[0114] The planning function 17b plans the position, orientation, and size of the FOV of the scan ROI based on the organ segmentation results obtained by the segmentation unit 110.

[0115] The rendering function 17c performs three-dimensional morphological rendering of the organ based on the organ division results obtained by the division unit 110.

[0116] For example, when abdominal magnetic resonance imaging is performed, the processing circuit 17 automatically creates a scan plan for scanning multiple organs in the abdomen using the positioning function 17a and the planning function 17b. Also, when abdominal magnetic resonance imaging is performed, the processing circuit 17 uses the rendering function 17c to render the three-dimensional morphology of the organs in the abdomen.

[0117] As described above, the magnetic resonance imaging apparatus 200 according to the fourth embodiment includes an image division device described in any of the embodiments and examples described above, and a positioning unit that positions the organ based on the division results of the organ. With this configuration, rapid positioning can be achieved, asymmetric structures can be detected, divided, and positioned without being limited by structural symmetry, characteristic information such as the position and size of multiple organs in three-dimensional space and accurate positioning can be provided, and the performance of the magnetic resonance imaging apparatus can be improved.

[0118] Furthermore, the magnetic resonance imaging apparatus 200 includes an image segmentation device described in any of the embodiments and examples described above, and a planning unit that plans the position, direction, and size of the scan ROI and FOV based on the organ segmentation results. With this configuration, it is possible to automate the scanning plan for multiple abdominal organs, standardize and improve the accuracy of abdominal scanning, simplify the operation flow and reduce the time required, achieve highly efficient automated scanning planning and high-performance automated scanning, and improve the performance of the magnetic resonance imaging apparatus.

[0119] Furthermore, the magnetic resonance imaging apparatus 200 includes an image division device described in any of the embodiments and examples described above, and a rendering unit that performs three-dimensional morphological rendering of the organ based on the division results of the organ. With this configuration, three-dimensional morphological rendering can be achieved, and the performance of the magnetic resonance imaging apparatus can be improved.

[0120] Furthermore, the magnetic resonance imaging apparatus 200 described above includes an image splitting device as described in any of the embodiments and examples described above, and therefore achieves the same technical effects as those described in the embodiments and examples described above.

[0121] The embodiments and examples of the technology disclosed in this application have been described above. In the embodiments and examples described above, the positioning image acquisition unit 101, the temporary positioning unit 102, the division unit 110, the optimization unit 103, and the body contour detection unit 104 are realized by a processing circuit such as a processor. In this case, the processing functions of the positioning image acquisition unit 101, the temporary positioning unit 102, the division unit 110, the optimization unit 103, and the body contour detection unit 104 are stored in a storage unit in the form of a program that can be executed by a computer, for example. The processing circuit then realizes the processing function corresponding to each processing unit by reading and executing each program stored in the storage unit. In other words, the image division device in the embodiments and examples described above has the processing functions shown in Figures 1, 6, and 11 when the processing circuit has read each program. In this case, the processing of each step shown in Figures 2, 7, and 12 is realized by the processing circuit reading and executing the program corresponding to each processing function from the storage unit.

[0122] Furthermore, the processing circuit described above is not limited to one implemented by a single processor, but may be composed of a combination of multiple independent processors, with each processor executing a program to realize each processing function. Also, the processing functions of each processing unit may be implemented by appropriately distributing or integrating them across one or more processing circuits. Moreover, the processing functions of each processing unit may be implemented by hardware such as circuits, software alone, or a mixture of hardware and software. In addition, although an example in which the program corresponding to each processing function is stored in a single memory circuit has been described here, the embodiments are not limited to this. For example, the program corresponding to each processing function may be distributed and stored in multiple memory circuits, and each processing unit may read and execute each program from each memory circuit.

[0123] Furthermore, the term "processor" used in the above description refers to circuits such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), 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), and a Field Programmable Gate Array (FPGA)). Here, instead of storing the program in a memory circuit, the program may be directly incorporated into the processor's circuit. In this case, the processor realizes its function by reading and executing the program incorporated into the circuit. Moreover, each processor in this embodiment is not limited to being configured as a single circuit; multiple independent circuits may be combined to form a single processor and realize its function.

[0124] Here, the program executed by the processor is provided pre-installed in ROM (Read Only Memory) or memory circuits. Alternatively, this program may be provided as a file in an installable or executable format on a computer-readable, non-transient storage medium such as a CD (Compact Disk)-ROM, FD (Flexible Disk), CD-R (Recordable), or DVD (Digital Versatile Disk). Furthermore, this program may be stored on a computer connected to a network such as the Internet and provided or distributed by downloading it via the network. For example, this program consists of modules containing the processing functions described above. In actual hardware, the CPU reads the program from a storage medium such as ROM and executes it, loading each module onto the main memory and generating it in the main memory.

[0125] Furthermore, in the embodiments and examples described above, each component of each illustrated device is a functional concept and does 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, and all or part of them can be functionally or physically distributed or integrated in any unit according to various loads and usage conditions. Moreover, each processing function performed by each device can be realized in whole or in any part by a CPU and a program that is analyzed and executed by the CPU, or by hardware using wired logic.

[0126] Furthermore, among the processes described in the embodiments and examples above, 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 by known methods. In addition, the processing procedures, control procedures, specific names, and information including various data and parameters shown in the above document and drawings can be arbitrarily changed unless otherwise specified.

[0127] The various types of data discussed in this specification are typically digital data.

[0128] According to at least one embodiment described above, rapid segmentation of organs in positioning images can be achieved.

[0129] While several embodiments have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be implemented in a variety of other forms, and various omissions, substitutions, modifications, and combinations of embodiments are possible without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of Symbols]

[0130] 100, 100a, 100b Image splitting device 101 Positioning image acquisition unit 102 Temporary positioning section 110 Split section 103 Optimization Unit 104 Body contour detection unit 200 Magnetic Resonance Imaging System

Claims

1. An image splitting device for magnetic resonance imaging, An acquisition unit that acquires a positioning image including a three-dimensional image including organs or a multi-layered two-dimensional image including the organs, Based on the positioning image, a determination unit obtains a plurality of two-dimensional cross-sectional images corresponding to each of a plurality of slices arranged in the layer direction within the spatial region of the positioning image, identifies a first two-dimensional cross-sectional image in which one end of the organ in the layer direction exists and a second two-dimensional cross-sectional image in which the other end of the organ in the layer direction exists from among the plurality of two-dimensional cross-sectional images, and determines the segment defined by the position corresponding to the first two-dimensional cross-sectional image and the position corresponding to the second two-dimensional cross-sectional image as the segment in which the organ exists in the layer direction. A division unit performs image division processing on the region of the positioning image in which the organ is located, and obtains the division result of the region of the organ. An image splitting device equipped with the following features.

2. The determination unit selects two or more two-dimensional cross-sectional images from the plurality of two-dimensional cross-sectional images based on the search algorithm, performs image segmentation on the selected two-dimensional cross-sectional images to identify a first two-dimensional cross-sectional image in which one end of the organ is located and a second two-dimensional cross-sectional image in which the other end of the organ is located. The image splitting device according to claim 1.

3. The determination unit selects the two or more cross-sectional two-dimensional images using an equally spaced selection method, a random selection method, or a selection method based on the distribution of the organs. The image splitting device according to claim 2.

4. The system further includes an optimization unit that, based on the results of segmenting the region of the organ, more precisely detects and optimizes local features that are any of a plurality of ends defining the positional range of the organ. The image splitting device according to claim 1.

5. The optimization unit selects the region of the local feature as the local feature region based on the division result of the organ region, and calculates and optimizes the central position of the local feature of the organ by performing three-dimensional surface detection or two-dimensional edge detection based on the selected local feature region. The image splitting device according to claim 4.

6. The local characteristic of the organ is one of the six ends that define the positional range of the organ. The image splitting device according to claim 4.

7. The aforementioned layer direction is either the head-to-foot direction, the front-to-back direction, or the left-to-right direction. The image splitting device according to claim 1.

8. The results of segmenting the organ region represent data that shows the outline, size, and location of the organ. The image splitting device according to claim 1.

9. The search algorithm is one of the following: linear search, binary search, tree-based search, or hash search. The image splitting device according to claim 3.

10. The aforementioned image segmentation process applies an image segmentation algorithm or deep learning. The image splitting device according to claim 1.

11. The system further includes a detection unit that detects the body surface area in a cross-sectional two-dimensional image based on a plurality of cross-sectional two-dimensional images acquired based on the positioning image, The optimization unit optimizes the local characteristics of the organ based on the division results of the body surface area and the organ region. The image splitting device according to claim 4.

12. The aforementioned organ is one of the following: liver, kidney, pancreas, spleen, or heart. The image splitting device according to any one of claims 1 to 11.

13. A method for image segmentation for magnetic resonance imaging, The steps include obtaining a three-dimensional image including organs or a positioning image including multiple layers of two-dimensional images including the organs, Based on the positioning image, a plurality of two-dimensional cross-sectional images are obtained corresponding to each of a plurality of slices arranged in the layer direction within the spatial region of the positioning image; a first two-dimensional cross-sectional image in which one end of the organ in the layer direction exists and a second two-dimensional cross-sectional image in which the other end of the organ in the layer direction exists are identified from among the plurality of two-dimensional cross-sectional images; and the segment defined by the position corresponding to the first two-dimensional cross-sectional image and the position corresponding to the second two-dimensional cross-sectional image is determined as the segment in which the organ exists in the layer direction. The steps include: performing image segmentation processing on the region of the positioning image in which the organ is located, and obtaining the segmentation result of the region of the organ; Image segmentation methods, including those mentioned above.

14. The image splitting device comprises the image splitting device according to any one of claims 1 to 11. Magnetic resonance imaging device.

15. The system further includes a positioning unit for positioning the organ based on the division results of the organ's region. The magnetic resonance imaging apparatus according to claim 14.

16. The system further includes a planning unit that plans the position, orientation, and size of the scan ROI (Region of Interest) and FOV (Field of View) based on the results of segmenting the organ region. The magnetic resonance imaging apparatus according to claim 14.

17. The system further includes a rendering unit that performs three-dimensional morphological rendering of the organ based on the results of segmenting the organ region. The magnetic resonance imaging apparatus according to claim 14.

Images