JP2016147045A - Image processing device, image processing method, image processing program, and magnetic resonance imaging device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing device capable of accurately and quickly performing correction of distortion of a DWI image.SOLUTION: An image processing device 100 includes a selection part 110, a first correction part 120, and a second correction part 130. The selection part 110 selects a diffusion weighted image that satisfies a predetermined condition from a plurality of diffusion weighted images whose movement detection gradient magnetic field pulse application directions or diffusion sensitivity coefficients are different. The first correction part 120 performs alignment between the selected diffusion weighted image and a reference image whose diffusion sensitivity coefficient is a reference value, and corrects the selected diffusion weighted image. The second correction part 130 performs alignment between the corrected diffusion weighted image and each of the other diffusion weighted images of the plurality of the diffusion weighted images, and corrects each of the other diffusion weighted images.SELECTED DRAWING: Figure 1

Description

  Embodiments described herein relate generally to medical image processing, and more specifically to an image processing apparatus, an image processing method, an image processing program, and a magnetic resonance imaging apparatus that perform correction on a diffusion weighted image.

  Diffusion tensor imaging (DTI) or diffusion tensor tractography (DTT) is an important means for performing functional analysis based on magnetic resonance (MR) images. Usually, a DTI image is obtained based on a DWI (diffusion-weighted imaging) image corresponding to a set (usually at least six) of different gradient directions (application directions).

  Since the DWI image may generate distortion (for example, image distortion caused by motion or eddy current), it is necessary to correct the DWI image. Usually, the DWI image is corrected by aligning each DWI image and diffusion sensitivity coefficient b = 0 image in a set (that is, in the same sequence) of DWI images.

“Fundamental study of distortion effects due to phase dispersion and eddy currents in diffusion-weighted images”, [August 3, 2015 search], Internet <URL: http://tohoku-b.umin.ac.jp/data/ 17bukaizassi / 17_page162.pdf>

  The problem to be solved by the present invention is to provide an image processing apparatus, an image processing method, an image processing program, and a magnetic resonance imaging apparatus capable of correcting distortion of a DWI image with accurate and fast processing.

  The image processing apparatus according to the embodiment includes a selection unit, a first correction unit, and a second correction unit. The selection unit selects a diffusion-weighted image satisfying a predetermined condition from a plurality of diffusion-weighted images having different motion detection gradient magnetic field pulse application directions or diffusion sensitivity coefficients. The first correction unit corrects the selected diffusion weighted image by aligning the selected diffusion weighted image with a reference image whose diffusion sensitivity coefficient is a reference value. The second correction unit performs alignment between the corrected diffusion weighted image and each of the other diffusion weighted images among the plurality of diffusion weighted images, and corrects each of the other diffusion weighted images.

FIG. 1 is a block diagram showing an arrangement example of a medical image processing apparatus according to an embodiment of the present invention. FIG. 2 is a block diagram showing an arrangement example of a medical image processing apparatus according to another embodiment of the present invention. FIG. 3 is a block diagram showing an arrangement example of the medical image processing apparatus according to another embodiment of the present invention. FIG. 4 is a block diagram showing an arrangement example of the medical image processing apparatus according to another embodiment of the present invention. FIG. 5 is a block diagram showing an arrangement example of a medical image processing apparatus according to another embodiment of the present invention. FIG. 6 is a block diagram showing an arrangement example of the magnetic resonance imaging apparatus according to one embodiment of the present invention. FIG. 7 is a flowchart showing a process of a medical image processing method according to an embodiment of the present invention. FIG. 8 is a flowchart illustrating a sub-process of a medical image processing method according to another embodiment of the present invention. FIG. 9 is a flowchart illustrating a sub-process of a medical image processing method according to another embodiment of the present invention. FIG. 10 is a flowchart illustrating a sub-process of a medical image processing method according to another embodiment of the present invention. FIG. 11 is a flowchart illustrating a process of a medical image processing method according to another embodiment of the present invention. FIG. 12 is a block diagram showing the configuration of a computer for realizing the method and apparatus of the present invention.

  The following outlines some embodiments of the present invention for a basic understanding of the present invention. This summary does not define key or critical areas of the invention and does not limit the scope of the invention. Its purpose is merely to be described in a simplified manner and for a more detailed description that follows.

  According to one embodiment of the present invention, the medical image processing apparatus includes a selection unit, a first correction unit, and a second correction unit. The selection unit selects at least one image with relatively high quality from a set of diffusion-weighted images based on a predetermined quality criterion, and sets it as a representative image. The first correction unit corrects the representative image by aligning the selected representative image and the diffusion sensitivity coefficient b = 0 image corresponding to the set of diffusion-weighted images. The second correction unit corrects the other images by aligning the other images of the set of diffusion-weighted images with the corrected representative image.

  According to another embodiment of the present invention, a magnetic resonance imaging apparatus including a medical image processing apparatus according to an embodiment of the present invention is provided.

  According to another embodiment of the present invention, a medical image processing method comprising: selecting at least one image having relatively good quality from a set of diffusion-weighted images based on a predetermined quality criterion. A step of making a representative image. The method further includes correcting the representative image by aligning the selected representative image with the diffusion sensitivity coefficient b = 0 image corresponding to the set of diffusion enhanced images. The method further includes correcting the other image by aligning another image of the set of diffusion-weighted images with the corrected representative image.

  Embodiments of the present invention are advantageous for improving correction accuracy and processing efficiency of DWI images.

  Hereinafter, by describing the present embodiment with reference to the drawings, it is possible to further understand the purpose, features, and merits of the present embodiment. The configuration in the drawings is merely for illustrating the principle of the present embodiment. In the drawings, the same or similar technical features or configurations are expressed using the same or similar drawing notations.

  Hereinafter, the present embodiment will be described with reference to the drawings. In the description, the configurations and features described in one drawing or one embodiment can be combined with the configurations and features shown in one or more other drawings or embodiments. Furthermore, for the sake of clarity, the description and description of the contents unrelated to the present embodiment and the configurations and processes well known to those skilled in the art are omitted in the drawings and description.

  As shown in FIG. 1, the medical image processing apparatus 100 according to an embodiment of the present invention includes a selection unit 110, a first correction unit 120, and a second correction unit 130. The medical image processing apparatus 100 is an example of an image processing apparatus.

  The selection unit 110 selects at least one image with relatively high quality from a set of diffusion-weighted images (DWI) based on a predetermined quality standard, and sets it as a representative image. That is, the selection unit 110 selects a diffusion-weighted image that satisfies a predetermined condition from a plurality of diffusion-weighted images having different diffusion sensitivity coefficients.

  In the present text, “a set of DWIs” is a DWI corresponding to a plurality of different diffusion sensitivity gradients in the same imaging region, and the diffusion sensitivity gradient field parameter is referred to as a diffusion sensitivity coefficient or b value. A set of DWIs is used to generate a DTI.

  The selection unit 110 can select a representative image based on a plurality of types of predetermined quality standards. For example, according to one embodiment, the selection unit can select the representative image based on the distortion degree and / or the configuration sharpness of the diffusion weighted image. That is, the selection unit 110 performs selection based on the degree of distortion and the definition of each of the plurality of diffusion weighted images as a predetermined condition. However, the predetermined quality standard is not limited to this. For example, as described below, one or a plurality of DWIs having a relatively high similarity to a corresponding b = 0 image in a set of DWIs are represented as representative images. Can be selected.

  The first correction unit 120 aligns the representative image selected by the selection unit 110 with an image corresponding to the diffusion sensitivity coefficient b = 0 of the set of diffusion-weighted images (hereinafter referred to as b = 0 image). The representative image can be corrected. In other words, the first correction unit 120 corrects the selected diffusion weighted image by aligning the selected diffusion weighted image with the reference image whose diffusion sensitivity coefficient is the reference value. For example, the first correction unit 120 corrects the selected diffusion weighted image using an image having the diffusion sensitivity coefficient of 0 as the reference image.

  Image correction based on alignment can be performed by using an existing method in this field. For example, Comprehensive Approach for Correction of Motion and Distribution in Diffusion-Weighted MRI, G.M. K. Rohde, et al. , Magnetic Resonance in Medicine 51: 103-114 (2004).

  When the number of representative images is two or more, a corresponding corrected representative image can be obtained by aligning each representative image with the b = 0 image.

  The second correction unit 130 corrects other images of the set of diffusion-weighted images by aligning the other images of the set of diffusion-weighted images with the corrected representative image. That is, the 2nd correction | amendment part 130 corrects each of the other diffusion emphasis images by aligning the corrected diffusion emphasis image with each of the other diffusion emphasis images among the plurality of diffusion emphasis images. The second correction unit 130 may employ a registration correction method similar to that of the first correction unit 120, or may employ a registration correction method different from that of the first correction unit 120.

  Further, when the number of representative images is two or more, the DWI can be corrected based on the alignment result of each of the DWIs in the set and the corrected representative images. A corrected representative image having a relatively low alignment deviation can be selected and corrected as a final alignment target, and correction can be performed by comprehensively considering the alignment results with a plurality of corrected representative images. Alternatively, the image other than the representative image is divided into a set corresponding to each representative image, and the correction is performed by aligning each set image and the corresponding corrected representative image. For example, grouping can be performed based on the similarity with the corrected representative image, and the second correction unit performs an alignment process on an image having a relatively high similarity.

  As described above, according to the existing method, the DWI image is corrected by aligning each DWI image and the b = 0 image. According to the embodiment of the present invention, a representative image selected from a set of DWI and a b = 0 image are aligned to correct the representative image, and the other DWI and the corrected representative image are aligned. Correct other images. Since the selected representative image has a relatively good image quality, it is usually relatively high in alignment accuracy and processing efficiency with the b = 0 image. In addition, since characteristics such as gray level distribution between DWI images are relatively similar, alignment between other images and the corrected representative image may have relatively high accuracy and processing efficiency. Is possible. Therefore, compared with the existing method, the medical image processing apparatus according to the embodiment of the present invention can correct a DWI image with higher accuracy and processing efficiency.

  The predetermined quality criterion with which the selection unit selects the representative image may include a DWI distortion degree, a configuration sharpness, or a combination of both. Next, each embodiment will be described in which a representative image is selected using the constituent definition and the degree of distortion as a predetermined quality standard.

  As shown in FIG. 2, the medical image processing apparatus 200 according to an embodiment of the present invention includes a selection unit 210, a first correction unit 220, a second correction unit 230, a determination unit 240, and a configuration definition estimation unit 250. The first correction unit 220 and the second correction unit 230 are the same as the corresponding configurations described in FIG.

  According to the present embodiment, the determination unit 240 and the configuration definition estimation unit 250 estimate the configuration definition of the DWI, and the selection unit 210 represents the representative image based on the configuration definition estimated by the configuration definition estimation unit 250. Select.

  Specifically, the determination unit 240 determines an average image of at least some of the DWIs in the set. The configuration definition estimation unit 250 estimates the configuration definition of the DWI based on the average image determined as DWI.

  In a certain DWI, the configuration information of the certain region may be lost due to causes such as eddy current. For example, the contrast of a certain region may be relatively poor, and the configuration characteristics of the region cannot be reflected. If the configuration information is lost in different DWIs, it may appear in different areas, so the effect of losing information in individual images on the average image is mitigated to other images there is a possibility. A DWI with a relatively small difference from the average image has good compositional sharpness.

  The average image may be determined based on all DWIs in the set of DWIs, or the average image may be determined based on some DWIs. Some of the DWIs may be selected randomly or may be selected based on predetermined criteria. For example, specific pre-processing (for example, a method based on texture or energy gradient) is adopted to eliminate DWI having a relatively large configuration information loss area and avoid its influence on the average image.

  In addition, the configuration definition of the DWI can be estimated based on a certain DWI and an average image by a plurality of methods. According to a specific embodiment, the configuration sharpness of the DWI can be estimated based on a sum of squared difference (SSD) between the DWI and the average image.

  In the embodiment described above, the average image of the DWI is determined and compared with the average image to estimate the configuration definition of the DWI. In addition, by adopting other image definition evaluation methods, it is possible to estimate the DWI image definition even when the average image cannot be acquired.

  For example, as illustrated in FIG. 2, the medical image processing apparatus according to the embodiment may not include the determination unit. The first estimation unit can estimate the configuration definition of the DWI based on the local dispersion diagram of the DWI. For example, based on a comparison between the local dispersion diagram of the DWI and the local dispersion diagram of the b = 0 image, the configuration definition of the DWI can be estimated. By comparing the high frequency regions of the images based on the local dispersion diagram with the b = 0 image as a reference, it is possible to estimate the DWI configuration definition with relatively high processing efficiency.

  In addition, other image definition evaluation methods existing in this field, for example, a texture-based method and an energy gradient-based method (for example, “Wavelet transform image definition evaluation method research based on texture analysis”, Liu Xingho Sei, << Chinese Instrument Academic Journal >>, refer to 2007, 8th period). It is possible to evaluate the structural definition of the DWI without comparing it with the average image or the b = 0 image, and select a representative image based on this evaluation for each DWI.

  Next, an embodiment in which a representative image is selected using the degree of distortion as a predetermined quality standard will be described.

  As illustrated in FIG. 3, the medical image processing apparatus 300 according to an embodiment includes a selection unit 310, a first correction unit 320, a second correction unit 330, a determination unit 340, and a distortion degree estimation unit 350. The first correction unit 320 and the second correction unit 330 are the same as the corresponding configurations described in FIG. The determination unit 340 and the distortion degree estimation unit 350 estimate the DWI distortion degree, and the selection unit 310 selects a representative image based on the distortion degree estimated by the distortion degree estimation unit 350.

  Specifically, the determination unit 340 determines an average image of at least some of the DWIs in the set of DWIs.

  In a certain DWI, distortion is generated due to movement or the like, and the distortion can include, for example, a translation component and a rotation component. By determining the average image, the influence of the distortion of the individual images can be alleviated to some extent. Therefore, a DWI having a relatively small difference from the average image has a relatively low degree of distortion.

  The distortion degree estimation unit 350 estimates the distortion degree of the DWI based on the DWI and the average image determined by the determination unit 340.

  Specifically, the distortion degree estimation unit 350 can estimate an offset component of the distortion of the DWI based on a positional difference between the centroid point of the DWI and the centroid of the average image. The rotational component of the DWI distortion can also be estimated based on the angular difference between the principal axis of the DWI and the principal axis of the average image.

  As described above, the embodiment in which the representative image is selected using the configuration definition and the degree of distortion as the predetermined quality criterion is described. However, the predetermined quality criterion for selecting the representative image is not limited to this. Next, an embodiment in which a representative image is selected as a predetermined quality standard based on the similarity between the DWI and the b = 0 image will be described with reference to FIG.

  As illustrated in FIG. 4, the medical image processing apparatus 400 according to an embodiment includes a selection unit 410, a first correction unit 420, a second correction unit 430, and a determination unit 440. The arrangement of the first correction unit 420 and the second correction unit 430 is similar to the corresponding unit.

  The determination unit 440 determines the degree of similarity between the DWI and the b = 0 image. The selection unit 410 selects a representative image based on the similarity determined by the determination unit 440.

  The determination unit 440 may determine the similarity between the DWI and the b = 0 image using a plurality of types of image similarity evaluation methods existing in the field, for example, a method based on a gray level histogram. .

  A DWI similar to a b = 0 image usually has a relatively good compositional sharpness and a relatively low degree of distortion. Furthermore, a more accurate correction result can be obtained by aligning a B = 0 image and a similar DWI.

  The exemplary embodiment in which the representative image is selected based on the image quality, the degree of distortion, and the similarity with the b = 0 image as the predetermined quality criterion will be described above. In addition, the representative image can be selected by a combination form of the above embodiments. For example, the overall image quality estimation is determined based on the configuration definition and the degree of distortion, and the representative image is selected based on the overall image quality estimation.

  As described above, an embodiment of the medical image processing apparatus for correcting the DWI image described above provides the DWI corrected by the medical image processing apparatus to other apparatuses and is used for post-processing, for example, generates a DTI. . The present invention also includes a medical image processing apparatus that can generate a DTI.

  As illustrated in FIG. 5, the medical image processing apparatus 500 includes a selection unit 510, a first correction unit 520, a second correction unit 530, and a generation unit 540. The arrangement of the selection unit 510, the first correction unit 520, and the second correction unit 530 can be similar to the corresponding units in the embodiments described above. The generation unit 540 generates a DTI based on the corrected set of DWI. A method of generating a DTI based on a set of DWIs is existing in this field, and will not be described here again.

  The embodiment of the present invention further includes a magnetic resonance imaging apparatus. As shown in FIG. 6, the magnetic resonance imaging apparatus 600 includes a medical image processing apparatus 610. The medical image processing apparatus 610 can have, for example, the arrangement of each embodiment described above with reference to FIGS. 1 to 5 or any combination thereof.

  In the above embodiment, a case has been described in which the selection units 110, 210, 310, 410, and 510 select a diffusion weighted image satisfying a predetermined condition from a plurality of DWIs having different b values. It is not limited to. For example, the selection units 110, 210, 310, 410, and 510 may select a diffusion weighted image satisfying a predetermined condition from a plurality of DWIs having different application directions of motion detection gradient magnetic field (MPG) pulses. Good. That is, the selection units 110, 210, 310, 410, 510 may select a diffusion-weighted image that satisfies a predetermined condition from a plurality of diffusion-weighted images having different MPG pulse application directions or diffusion sensitivity coefficients (b values). Good.

  As described above, a certain process or method is further developed in the description process of the medical image processing apparatus in the embodiment. Hereinafter, a method that is a case where certain details described above do not overlap will be described.

  As shown in FIG. 7, according to the medical image processing method of one embodiment, in step S710, a diffusion-weighted image that satisfies a predetermined condition is selected from a plurality of diffusion-weighted images. For example, at least one image with relatively high quality is selected from a set of DWIs based on a predetermined quality standard, and is set as a representative image.

  A representative image can be selected based on a plurality of predetermined quality criteria. For example, according to one embodiment, a representative image can be selected based on the degree of distortion and / or configuration sharpness of the DWI. Further, for example, one or a plurality of DWIs having a relatively high similarity to the corresponding b = 0 image in a set of DWIs can be selected as the representative image.

  In step S720, the selected diffusion weighted image is aligned with the reference image to correct the selected diffusion weighted image. For example, the representative image is corrected by aligning the selected representative image with the b = 0 image corresponding to the set of DWI.

  As described above, the representative image and the b = 0 image can be aligned by a plurality of types of methods.

  When the number of representative images is two or more, a corresponding corrected representative image can be obtained by aligning each representative image with the b = 0 image.

  In step S730, the corrected diffusion weighted image is aligned with each of the other diffusion weighted images among the plurality of diffusion weighted images to correct each of the other diffusion weighted images. For example, by aligning another image in the DWI of the set with the corrected representative image, the other image is corrected.

  When the number of representative images is two or more, the DWI can be corrected based on the alignment result of each of the DWIs in the set and the corrected representative image. Alternatively, another image other than the representative image is divided into a set corresponding to each representative image, and the correction is performed by aligning each set image and the corresponding corrected representative image.

  According to the embodiment of the present invention, a representative image selected from a set of DWI and a b = 0 image are aligned to correct the representative image, and the other DWI and the corrected representative image are aligned. Correct other images. Compared with the existing method, the medical image processing apparatus according to the embodiment of the present invention can correct the DWI image with higher accuracy and processing speed.

  In addition, for example, since the signal value of a DWI image captured with a high b value is small and the difference between a b = 0 image and a DWI image is large, the accuracy of distortion correction has been lowered. However, the medical image processing apparatus according to the embodiment selects a DWI image suitable for alignment with the b = 0 image. Then, the selected DWI image and the b = 0 image are aligned to correct the DWI image, and the remaining DWI image is corrected using the corrected DWI image. Therefore, the medical image processing apparatus according to the embodiment can accurately perform distortion correction even for an image having a high b value.

  Subsequently, embodiments in which a representative image is selected using the definition and the degree of distortion as a predetermined quality criterion will be described.

  FIG. 8 shows an example process of selecting a representative image in the medical image processing method according to the embodiment. A representative image is selected using the structural definition as a predetermined quality standard.

  In step S810, an average image of at least some of the DWIs in the set is determined.

  An average image can be determined based on all or some of the DWIs in a set of DWIs. Some DWIs for determining an average image can be selected randomly or based on a predetermined criterion (eg, a relatively small information loss area).

  In step S820, the configuration definition of each DWI is estimated based on each DWI and the average image.

  As described above, a DWI having a relatively small difference from the average image has a relatively good configuration definition, and the difference between the DWI and the average image can be evaluated by a plurality of methods.

  For example, according to one specific embodiment, the configuration sharpness of the DWI can be estimated based on the sum of squared differences (SSD) between the DWI and the average image.

  In the above embodiment, an average image of the DWI is determined, and the configuration definition of the DWI is estimated by comparing with the average image. Further, it is possible to estimate the image definition of the DWI when the average image is not taken.

  According to one embodiment, the configuration sharpness of the DWI is estimated based on the local dispersion map of the DWI. For example, based on a comparison between the local dispersion diagram of the DWI and the local dispersion diagram of the b = 0 image, the configuration definition of the DWI can be estimated.

  In addition, other image definition evaluation methods existing in this field can be employed, and for example, the above-described texture-based method and energy gradient-based method can be employed.

  FIG. 9 shows a process example of selecting a representative image in a medical image processing method according to another embodiment. A representative image is selected using the degree of distortion as a predetermined quality standard.

  As shown in FIG. 9, in step S910, an average image of at least some of the DWIs in the set of DWIs is determined. An average image can be determined by employing a method similar to the method described with reference to FIG.

  In step S920, the degree of distortion of each DWI is estimated based on each DWI and the average image.

  According to a specific embodiment, the distortion degree estimation process may include an offset component and rotation component estimation process. FIG. 10 shows an example of a sub-process for estimating the degree of distortion of the DWI based on the DWI and the average image.

  In step S1010, the offset component of the distortion of each DWI is estimated based on the position difference between the center of gravity of each DWI and the center of gravity of the average image.

  In step S1020, the rotational component of the distortion of each DWI is estimated based on the angle difference between the main axis of each DWI and the main axis of the average image.

  In the above, an embodiment in which a representative image is selected using the configuration definition and the distortion degree as a predetermined quality standard will be described. According to another embodiment, the step of selecting the representative image based on the predetermined quality criterion determines the similarity between the DWI and the b = 0 image, and selects the representative image based on the determined similarity. Can include.

  In addition, the representative image can be selected by a combination method of the above methods. For example, the overall image quality estimation is determined based on the configuration definition and the degree of distortion, and the representative image is selected based on the overall image quality estimation.

  In the above, an embodiment of a medical image processing method for correcting a DWI image has been described. The present invention also includes a medical image processing method for generating a DTI.

  As shown in FIG. 11, in the medical image processing method according to the present embodiment, steps S1110 to S1130 are similar to the corresponding steps described above. That is, in step S1110, a diffusion weighted image satisfying a predetermined condition is selected from a plurality of diffusion weighted images, and in step S1120, the selected diffusion weighted image and the reference image are aligned, and the selected diffusion weighted image is selected. The image is corrected, and in step S1130, the corrected diffusion weighted image is aligned with each of the other diffusion weighted images among the plurality of diffusion weighted images to correct each of the other diffusion weighted images.

  Further, the medical image processing method according to the present embodiment includes step S1140 of generating a diffusion tensor image (DTI) based on a plurality of corrected diffusion weighted images (DWI).

  As an example, each step of the medical image processing method and each configuration and / or part of the medical image processing apparatus may be implemented as software, firmware, hardware, or a combination thereof. When implemented via software or firmware, in order to implement the software program (image processing program) of the above method, a computer with a dedicated hardware structure (for example, as shown in FIG. 12) from a memory medium or via a network General-purpose computer 1200) can be downloaded and configured, and various functions can be implemented with various programs downloaded to the computer.

  In FIG. 12, an arithmetic processing unit (ie, CPU) 1201 is a program stored in a read-only memory (ROM) 1202 or a program written from a storage unit 1208 to a read / write memory (RAM) 1203. Based on this, various processes are performed. The RAM 1203 also stores data necessary when the CPU 1201 performs various processes and the like as necessary. The CPU 1201, ROM 1202, and RAM 1203 are connected to each other via a combination line 1204. An input / output interface 1205 is also connected to the combined line 1204.

  The following units are connected to an input / output interface 1205: an input unit 1206 (including a keyboard and a mouse), an output unit 1207 (a monitor such as a cathode ray tube (CRT), a liquid crystal monitor (LCD), etc., and a speaker. Storage unit 1208 (including a hard disk), communication unit 1209 (network interface card such as a LAN card, modem, etc.). The communication unit 1209 performs communication processing via a network (for example, the Internet). The driver 1210 can also be connected to the input / output interface 1205 as needed. The removable medium 1211 is, for example, a magnetic disk, an optical disk, an MO, a semiconductor memory, or the like. The removable medium 1211 is attached to the driver 1210 as necessary, reads out a computer program as necessary, and is downloaded to the storage unit 1208. .

  When the above-described system processing is performed via software, a program constituting the software may be downloaded from a network (for example, the Internet) or a storage medium (for example, removable medium 1211).

  For those skilled in the art, the storage medium storing such a program as shown in FIG. 12 is not limited to the removable medium 1211 that provides the user with the program from a location apart from the apparatus. Examples of the removable medium 1211 include a magnetic disk (floppy (registered trademark) disk), an optical disk (including CD-ROM and DVD), a magnetic optical disk (including MiniDisc (MD, registered trademark)), and a semiconductor memory. Including. Further, the storage medium may be the ROM 1202, or a form in which a program is stored therein such as a hard disk included in the storage unit 1208, and the program is sent to the user from a device including them.

  The present invention can also be applied to a program product that stores a device-readable command code as a memory. When the command code is read through the device, the image processing method of the present invention is implemented.

  A storage medium for accepting a program product storing the device-readable command code can also be applied to the present invention. The storage medium is not limited to a hard disk, an optical disk, a magnetic optical disk, a memory card, or a memory stick.

  In the specific embodiment described above, the same method is applied to one or more other implementation methods for the features shown in one embodiment, combined with other implementation methods, or other implementations. It is also possible to switch to features in the method.

  Furthermore, when the term “include / include” is used, it indicates the presence of a feature / configuration / step or structure. However, it does not mean the presence or addition of other features, configurations, steps or structures.

  In the above-described embodiment, each step or configuration is indicated using a figure number symbol having a numerical configuration. However, it should be understood by those skilled in the art that these figure number symbols are merely for the convenience of explanation and drawing and do not represent the order or any other limitations.

  In addition, the method according to the present embodiment is not limited to being performed along the time order described in the detailed description column, and may be performed simultaneously or independently along other time orders. good. Therefore, the execution order of the method described in the detailed description of the present application does not limit the configuration of the technical scope of the present embodiment.

  In the above description, the present embodiment has already been described with the description of the specific embodiment of the present embodiment. However, all the above-described embodiments are merely examples, and are not intended to be limiting. Those skilled in the art can make various modifications, improvements, or equivalent designs of the present embodiment within the spirit and scope of the claims. These modifications, improvements, or equivalents are included within the protection scope of the present embodiment.

DESCRIPTION OF SYMBOLS 100 Medical image processing apparatus 110 Selection part 120 1st correction | amendment part 130 2nd correction | amendment part

Claims (13)

A selection unit that selects a diffusion-weighted image satisfying a predetermined condition from a plurality of diffusion-weighted images having different application directions of motion detection gradient magnetic field pulses or diffusion sensitivity coefficients;
A first correction unit that performs alignment between the selected diffusion-weighted image and a reference image whose diffusion sensitivity coefficient is a reference value, and corrects the selected diffusion-weighted image;
A second correction unit that performs alignment between the corrected diffusion-weighted image and each of the other diffusion-weighted images among the plurality of diffusion-weighted images and corrects each of the other diffusion-weighted images. Image processing device. The first correction unit corrects the selected diffusion weighted image using an image having a diffusion sensitivity coefficient of 0 as the reference image.
The image processing apparatus according to claim 1. The selection unit selects, as the predetermined condition, based on a degree of distortion and a configuration sharpness of each of the plurality of diffusion weighted images.
The image processing apparatus according to claim 1. A determining unit that determines an average image of at least some of the plurality of diffusion-weighted images;
The image processing apparatus according to claim 3, further comprising: a configuration definition estimation unit that estimates a configuration definition of each diffusion enhancement image based on each of the plurality of diffusion enhancement images and the average image. The configuration definition estimation unit estimates the configuration definition of each diffusion weighted image based on the sum of squared differences between each diffusion weighted image and the average image.
The image processing apparatus according to claim 4. The image processing apparatus according to claim 2, further comprising: a configuration definition estimation unit that estimates a configuration definition of each diffusion weighted image based on a local variogram of each of the plurality of diffusion weighted images. A determining unit that determines an average image of at least some of the plurality of diffusion-weighted images;
The image processing apparatus according to claim 2, further comprising: a distortion degree estimation unit that estimates a distortion degree of each diffusion weighted image based on each of the plurality of diffusion weighted images and the average image. The distortion degree estimation unit
Based on the position difference between the centroid point of each of the plurality of diffusion-weighted images and the centroid point of the average image, the offset component of the degree of distortion of each diffusion-weighted image is estimated,
Estimating a rotational component of the degree of distortion of each diffusion-weighted image based on the difference in angle between the main axis of each of the plurality of diffusion-weighted images and the main axis of the average image;
The image processing apparatus according to claim 7. A determination unit for determining a similarity between each of the plurality of diffusion-weighted images and the reference image;
The selection unit selects as the predetermined condition based on the similarity.
The image processing apparatus according to claim 1. A generator that generates a diffusion tensor image using the plurality of diffusion-weighted images corrected by the first correction unit and the second correction unit;
The image processing apparatus according to claim 1. The image processing apparatus according to claim 1 is provided.
Magnetic resonance imaging device. Select a diffusion-weighted image satisfying a predetermined condition from a plurality of diffusion-weighted images with different motion detection gradient magnetic field pulse application directions or diffusion sensitivity coefficients,
The selected diffusion weighted image is aligned with the reference image whose diffusion sensitivity coefficient is a reference value to correct the selected diffusion weighted image,
An image processing method comprising: aligning the corrected diffusion weighted image with each of the other diffusion weighted images of the plurality of diffusion weighted images and correcting each of the other diffusion weighted images. Select a diffusion-weighted image satisfying a predetermined condition from a plurality of diffusion-weighted images with different motion detection gradient magnetic field pulse application directions or diffusion sensitivity coefficients,
The selected diffusion weighted image is aligned with the reference image whose diffusion sensitivity coefficient is a reference value to correct the selected diffusion weighted image,
Aligning the corrected diffusion-weighted image with each of the other diffusion-weighted images of the plurality of diffusion-weighted images, and causing the computer to execute each process for correcting each of the other diffusion-weighted images. Image processing program.

Images