JP5479115B2 - Image processing apparatus and magnetic resonance imaging apparatus - Google Patents

Description

  The present invention relates to an image processing apparatus and a magnetic resonance imaging apparatus.

  Conventionally, as one of image diagnostic apparatuses used in the medical field, there is a magnetic resonance imaging (MRI) apparatus. The MRI apparatus magnetically excites a nuclear spin of a subject placed in a static magnetic field with an RF (Radio Frequency) signal having a Larmor frequency, and reconstructs an image from an MR signal generated by the excitation.

  One of imaging methods using such an MRI apparatus is diffusion imaging. Diffusion imaging is a technique for imaging a diffusion weighted image (DWI) that emphasizes the diffusion effect in which particles such as water molecules are scattered by Brownian motion due to heat. This diffusion imaging has attracted attention because it is useful for early diagnosis of cerebral infarction. Diffusion imaging has been developed as Diffusion Tensor Imaging (DTI) in the cranial nerve region where anisotropy of nerve fibers is detected and, conversely, nerve fibers are depicted using anisotropy. ing.

  In recent years, an imaging technique called diffusion tensor tractography (DTT) has been attracting attention (see, for example, Non-Patent Document 1). The DTT is a diffusion tensor tractography image (hereinafter referred to as a DTT image) in which a maximum diffusion direction at an arbitrary pixel is traced and a trajectory of a diffusion tensor image (hereinafter referred to as a DTI image) obtained by the DTI is traced. ).

  In such a DTT, a calculation start region (also referred to as a region of interest (ROI)) is usually set on the DWI image or DTI image by the operator of the MRI apparatus when the DTT image is generated. This calculation start area is set in order to specify the position of the pixel from which tracking in the maximum diffusion direction is started. When an operator such as a doctor or an engineer sets a DTT calculation start area, the operator determines the position of the calculation start area based on the DWI image or information displayed on the DTI image.

Susumu Mori and Jiangyang Zhang, "Principles of Diffusion Tensor Imaging and Its Applications to Basic Neuroscience Research", Neuron 51, p.527-539, September 7, 2006

  However, in the conventional technique described above, there are cases where a DTT image useful in the clinic cannot be generated.

  Specifically, in clinical practice, not only DWI images and DTI images but also various medical images suitable for treatment and diagnosis are referred to. For example, as a clinical reference image, an fMRI (Functional MRI) image, a PWI (Perfusion Weighted Imaging) image, or the like is used. However, as described above, in the conventional technique, the DTT calculation start area is usually set using a DWI image or a DTI image. Therefore, when a DWI image or a DTI image is not used as a reference image, the operator cannot set a DTT calculation start area based on information on an image suitable for clinical use. As described above, there are cases where the conventional technique cannot generate a clinically useful DTT image.

  This problem is not limited to diffusion imaging using an MRI apparatus, but other medical image diagnostic apparatuses such as an X-ray CT (Computed Tomography) apparatus and a PET (Positron Emission Tomography) apparatus, an image processing apparatus, and the like. This is a problem that similarly occurs in an apparatus that processes an image generated by diffusion imaging.

  The present invention has been made in view of the above, and an object thereof is to provide an image processing apparatus and a magnetic resonance imaging apparatus capable of generating a DTT image useful in clinical practice.

In order to solve the above-described problems and achieve the object, an image processing apparatus according to an aspect of the present invention extracts a lesion area on a clinical reference image , and then extracts the lesion area by a predetermined size. A region-of-interest setting unit that generates an expanded margin region and sets a region outside the lesion region in the margin region as a region of interest ; a diffusion tensor image representing anisotropy of molecular diffusion in the subject; A calculation start area is set on the diffusion tensor image aligned with the reference image based on information relating to the alignment means for aligning the reference image and the region of interest set on the reference image, and the calculation And an image generation means for generating a diffusion tensor tract image based on the start area.
The image processing apparatus according to another aspect of the present invention also includes a region including a portion suspected of having a cerebral infarction when the brain DWI image and the PWI image are used as clinical reference images. Region-of-interest setting means that extracts from each image and sets the region where the extracted regions do not overlap as a region of interest, and aligns the diffusion tensor image that represents the anisotropy of molecular diffusion in the subject and the reference image And a calculation start region on the diffusion tensor image aligned with the reference image based on information on the region of interest set on the reference image, and a diffusion tensor based on the calculation start region An image generation means for generating a tract image is provided.

An image processing apparatus according to another aspect of the present invention generates a margin area obtained by extracting a lesion area on a clinical reference image and then expanding the lesion area by a predetermined size. A region-of-interest setting unit that sets a region outside the lesion region as a region of interest, and a positioning unit that aligns a diffusion tensor image representing anisotropy of molecular diffusion in the subject and the reference image A display control means for setting the region of interest set on the reference image on a diffusion tensor image aligned with the reference image, and displaying the region of interest on a display unit; Operation accepting means for accepting an operation for changing the region of interest from the operator, and alignment with the reference image based on information on the region of interest changed by accepting the operation. Setting the calculation start region on the diffusion tensor image, characterized in that an image generating means for generating a spread Ten Salt lacto image based on the calculation start region.
The image processing apparatus according to another aspect of the present invention also includes a region including a portion suspected of having a cerebral infarction when the brain DWI image and the PWI image are used as clinical reference images. Region-of-interest setting means that extracts from each image and sets the region where the extracted regions do not overlap as a region of interest, and aligns the diffusion tensor image that represents the anisotropy of molecular diffusion in the subject and the reference image Positioning means, a display control means for setting a region of interest set on the reference image on a diffusion tensor image aligned with the reference image, and displaying the region of interest on a display unit, and the display unit Operation receiving means for receiving an operation for changing the region of interest displayed on the operator from the operator, and information on the region of interest changed by receiving the operation Set the calculation start region on aligned diffusion tensor images in the reference image, characterized in that an image generating means for generating a spread Ten Salt lacto image based on the calculation start region.

In addition, a magnetic resonance imaging apparatus according to another aspect of the present invention is a first image that generates a diffusion tensor image representing anisotropy of molecular diffusion in a subject from data collected using a magnetic resonance phenomenon. After extracting the lesion area on the clinical reference image with the generation means, a margin area is generated by expanding the lesion area by a predetermined size, and the area outside the lesion area in the margin area A region of interest setting means for setting the region of interest, a positioning means for aligning the diffusion tensor image and the reference image, and a position on the reference image based on information on the region of interest set on the reference image Second image generation means for setting a calculation start region on the combined diffusion tensor image and generating a diffusion tensor tract image based on the calculation start region; Characterized by comprising.
In addition, a magnetic resonance imaging apparatus according to another aspect of the present invention is a first image that generates a diffusion tensor image representing anisotropy of molecular diffusion in a subject from data collected using a magnetic resonance phenomenon. When a brain DWI image and a PWI image are used as a generation means and a clinical reference image, a region including a portion suspected of having a cerebral infarction is extracted from each of the DWI image and the PWI image, and each extracted region is A region of interest setting means for setting a non-overlapping region as a region of interest, a positioning means for aligning the diffusion tensor image and the reference image, and the information on the region of interest set on the reference image A calculation start area is set on the diffusion tensor image aligned with the reference image, and the diffusion tensor tract image is set based on the calculation start area. Characterized in that a second image generating means for forming.

In addition, a magnetic resonance imaging apparatus according to another aspect of the present invention is a first image that generates a diffusion tensor image representing anisotropy of molecular diffusion in a subject from data collected using a magnetic resonance phenomenon. After extracting the lesion area on the clinical reference image with the generation means, a margin area is generated by expanding the lesion area by a predetermined size, and the area outside the lesion area in the margin area Region of interest setting means, positioning means for aligning the diffusion tensor image and the reference image, and diffusion tensor image in which the region of interest set on the reference image is aligned with the reference image Display control means for setting the area of interest to be displayed on the display unit, and operation reception for accepting an operation for changing the area of interest displayed on the display unit from the operator And a calculation start region is set on the diffusion tensor image aligned with the reference image based on the information on the region of interest changed by receiving the operation, and the diffusion tensor tract image is set based on the calculation start region. And a second image generation means for generating.
In addition, a magnetic resonance imaging apparatus according to another aspect of the present invention is a first image that generates a diffusion tensor image representing anisotropy of molecular diffusion in a subject from data collected using a magnetic resonance phenomenon. When a brain DWI image and a PWI image are used as a generation means and a clinical reference image, a region including a portion suspected of having a cerebral infarction is extracted from each of the DWI image and the PWI image, and each extracted region is A region-of-interest setting unit that sets a region that does not overlap as a region of interest, a positioning unit that aligns the diffusion tensor image and the reference image, and a region of interest set on the reference image is positioned in the reference image Set on the combined diffusion tensor image, display control means for displaying the region of interest on the display unit, and change the region of interest displayed on the display unit An operation accepting unit that accepts an operation from an operator, and a calculation start region is set on the diffusion tensor image that is aligned with the reference image based on information about the region of interest changed by the acceptance of the operation. And a second image generating means for generating a diffusion tensor tract image based on the above.

According to the present invention, it is possible to generate a clinically useful DTT image.

FIG. 1 is a diagram illustrating an overall configuration of the MRI apparatus according to the first embodiment. FIG. 2 is a functional block diagram illustrating the configuration of the storage unit and the data processing unit according to the first embodiment. FIG. 3 is a functional block diagram illustrating the configuration of the storage unit, the image processing unit, and the control unit according to the first embodiment. FIG. 4A is a diagram (1) for explaining setting of a region of interest when a DWI image of the head is used. FIG. 4B is a diagram (2) for explaining setting of a region of interest when a DWI image of the head is used. FIG. 4C is a diagram (3) for explaining setting of a region of interest when a DWI image of the head is used. FIG. 5 is a diagram for explaining setting of a region of interest when an fMRI image is used. FIG. 6 is a diagram for explaining setting of a region of interest using a PWI image and a DWI image of the head. FIG. 7 is a diagram for explaining the generation of the DTT image by the tractography generation unit according to the first embodiment. FIG. 8 is a diagram (1) illustrating an example of the DTT image displayed by the display control unit according to the first embodiment. FIG. 9 is a diagram (2) illustrating an example of the DTT image displayed by the display control unit according to the first embodiment. FIG. 10 is a flowchart illustrating a flow of creating a DTT image by the MRI apparatus according to the first embodiment. FIG. 11 is a functional block diagram illustrating the configuration of the storage unit, the image processing unit, and the control unit according to the second embodiment. FIG. 12 is a flowchart illustrating a flow of creating a DTT image by the MRI apparatus according to the second embodiment.

  Embodiments of an image processing apparatus and an MRI apparatus according to the present invention will be described below in detail with reference to the drawings. In addition, this invention is not limited by the Example shown below.

  In the embodiments described below, diffusion imaging is called DWI, and an image generated by DWI is called a DWI image. Further, diffusion tensor imaging is called DTI, and an image generated by DTI is called a DTI image. Also, diffusion tensor tractography is called DTT, and an image generated by DTT is called a DTT image. An image generated by fMRI is called an fMRI image, and an image generated by PWI is called a PWI image.

  First, the configuration of the MRI apparatus according to the first embodiment will be described. FIG. 1 is a diagram illustrating the overall configuration of the MRI apparatus 100 according to the first embodiment. As shown in FIG. 1, the MRI apparatus 100 includes a gantry unit 10, a gradient magnetic field power supply 20, an RF transmission unit 30, an RF reception unit 40, a sequence control unit 50, a bed unit 60, and a computer system 70. Have

  The gantry 10 irradiates a subject P placed in a static magnetic field with a high-frequency magnetic field, thereby detecting an MR signal emitted from the subject P. The gantry 10 includes a static magnetic field magnet 11, a gradient magnetic field coil 12, a transmission RF (Radio Frequency) coil 13, and a reception RF coil 14.

  The static magnetic field magnet 11 is formed in a hollow cylindrical shape, and generates a uniform static magnetic field in a space in the cylinder. For example, a permanent magnet or a superconducting magnet is used as the static magnetic field magnet 11.

  The gradient coil 12 is formed in a hollow cylindrical shape and is disposed inside the static magnetic field magnet 11. The gradient coil 12 has three coils corresponding to the X, Y, and Z axes orthogonal to each other. Each coil receives a current supply from a gradient magnetic field power source 20 described later, and generates a gradient magnetic field whose magnetic field intensity changes along each of the X, Y, and Z axes. The Z-axis direction is the same as the static magnetic field.

  The gradient magnetic fields of the X, Y, and Z axes generated by the gradient coil 12 correspond to, for example, the slice selection gradient magnetic field Gs, the phase encoding gradient magnetic field Ge, and the readout gradient magnetic field Gr, respectively. The slice selection gradient magnetic field Gs is used to arbitrarily determine an imaging section. The phase encoding gradient magnetic field Ge is used to change the phase of an echo signal (MR signal) in accordance with the spatial position. The readout gradient magnetic field Gr is used to change the frequency of the echo signal in accordance with the spatial position.

  The transmission RF coil 13 is arranged inside the gradient magnetic field coil 12 and receives a high frequency pulse from the RF transmission unit 30 to generate a high frequency magnetic field.

  The reception RF coil 14 is disposed inside the gradient magnetic field coil 12 and receives an echo signal emitted from the subject P due to the influence of the high-frequency magnetic field generated by the transmission RF coil 13. Then, the receiving RF coil 14 outputs the received echo signal to the RF receiving unit 40.

  The gradient magnetic field power supply 20 supplies a current to the gradient magnetic field coil 12. The RF transmitter 30 transmits a high-frequency pulse corresponding to the Larmor frequency to the transmission RF coil 13. The RF receiving unit 40 generates raw data by digitizing the MR signal output from the receiving RF coil 14, and transmits the generated raw data to the sequence control unit 50.

  The sequence control unit 50 scans the subject P by driving the gradient magnetic field power source 20, the RF transmission unit 30, and the RF reception unit 40 based on the sequence information transmitted from the computer system 70. Further, when the raw data is transmitted from the RF receiving unit 40 as a result of scanning the subject P, the sequence control unit 50 transfers the raw data to the computer system 70.

  The sequence information here refers to the strength of power supplied to the gradient magnetic field coil 12 by the sequence controller 50 and the timing of supplying power, the strength of the RF signal transmitted from the RF transmitter 30 to the transmission RF coil 136, and the like. This is information defining a procedure for performing scanning, such as a timing at which an RF signal is transmitted and a timing at which the RF receiver 40 detects an echo signal.

  The bed unit 60 includes a top plate on which the subject P is placed, and the subject P is inserted into the opening of the gantry unit 10 during imaging by moving the top plate.

  The computer system 70 is a device that performs overall control of the MRI apparatus 100, data collection, image reconstruction, and the like, and includes an interface unit 71, an input unit 72, a display unit 73, a storage unit 74, and a data processing unit 75. And an image processing unit 76 and a control unit 77.

  The interface unit 71 controls input / output of various signals exchanged with the sequence control unit 50. For example, the interface unit 71 transmits sequence information to the sequence control unit 50 and receives raw data from the sequence control unit 50. Further, when the raw data is received, the interface unit 71 stores the received raw data in the storage unit 74 for each subject P.

  The input unit 72 receives various instructions and information input from the operator. As the input unit 72, for example, a pointing device such as a mouse or a trackball, a selection device such as a mode change switch, or an input device such as a keyboard is used.

  The display unit 73 displays various images referred to by the operator and a GUI (Graphical User Interface) for receiving various operations from the operator. For example, a display device such as a liquid crystal monitor or a CRT monitor is used as the display unit 73.

  The storage unit 74 stores the raw data transmitted from the sequence control unit 50 and various image data generated by the data processing unit 75 and the image processing unit 76 described later for each subject P. The information stored in the storage unit 74 will be described in detail later.

  The data processing unit 75 reconstructs an image from the raw data stored in the storage unit 74. In addition, the data processing unit 75 generates a magnetic resonance image, a DWI image, or the like that shows morphological information in the body of the subject P. The functions of the data processing unit 75 will be described in detail later.

  The image processing unit 76 performs image processing such as threshold processing on the image data stored in the storage unit 74 to detect a calculation start region candidate used in DTT. In addition, the image processing unit 76 aligns a plurality of types of images stored in the storage unit 74. The functions of the image processing unit 76 will be described in detail later.

  The control unit 77 performs overall control of the MRI apparatus 100 by performing control movement between the above-described functional units and data transfer between the functional unit and the storage unit. Specifically, the control unit 77 includes a CPU (Central Processing Unit) and a memory, and controls each unit of the MRI apparatus 100 by executing various programs using them. For example, the control unit 77 generates sequence information based on the imaging conditions set by the operator, and transmits the generated sequence information to the sequence control unit 50 to perform various types of imaging. The functions of the control unit 77 will be described later in detail.

  The configuration of the MRI apparatus 100 according to the first embodiment has been described above. Under such a configuration, in the MRI apparatus 100 according to the first embodiment, the data processing unit 75 generates a DTI image representing the anisotropy of molecular diffusion in the subject. Further, the image processing unit 76 sets a region of interest on the reference image based on the image information included in the clinical reference image. The image processing unit 76 aligns the DTI image generated by the data processing unit 75 and the reference image. Then, the image processing unit 76 sets the region of interest set on the reference image as a calculation start region on the DTI image aligned with the reference image, and generates a DTT image based on the calculation start region. Therefore, according to the first embodiment, since a DTT image is generated based on a reference image suitable for clinical use, it is possible to generate a clinically useful DTT image.

  Next, the storage unit 74, the data processing unit 75, the image processing unit 76, and the control unit 77 will be described in detail with reference to FIGS. FIG. 2 is a functional block diagram illustrating configurations of the storage unit 74 and the data processing unit 75 according to the first embodiment. FIG. 3 is a functional block diagram illustrating configurations of the storage unit 74, the image processing unit 76, and the control unit 77 according to the first embodiment.

  As illustrated in FIG. 2, the storage unit 74 includes a raw data storage unit 74a, a reconstructed image storage unit 74b, and an analysis image storage unit 74c.

  The raw data storage unit 74a stores the raw data transmitted from the sequence control unit 50 for each subject. The reconstructed image storage unit 74b stores the reconstructed image generated by the image reconstructing unit 75a described later for each subject. The analysis image storage unit 74c stores various analysis images generated by an analysis image generation unit 75b described later for each subject.

  As illustrated in FIG. 2, the data processing unit 75 includes an image reconstruction unit 75a and an analysis image generation unit 75b.

  The image reconstruction unit 75a generates a reconstructed image by performing post-processing, that is, reconstruction processing such as Fourier transform processing, on the raw data stored in the raw data storage unit 74a. For example, the image reconstruction unit 75a generates a DWI image. Further, the image reconstruction unit 75a generates morphological images such as a T1 weighted image (T1 Weighted Image) and a T2 weighted image (T2 Weighted Image). Then, the image reconstruction unit 75a stores the generated reconstruction image in the reconstruction image storage unit 74b.

  Here, DWI will be described. In DWI, a pulse sequence with application of an MPG (Motion Probing Gradient) pulse that emphasizes attenuation of MR signals due to diffusion is used. Here, the signal intensity S due to diffusion is most simply expressed as the following equation (1).

S = PD (1-exp (-TR / T1) * exp (-TE / T2))
* Exp (-bD) (1)

Here, TE is echo time, TR is repetition time, and PD is proton density. T1 and T2 are signal relaxation times, and D is a diffusion coefficient (a diffusion coefficient representing the degree of diffusion). In addition, b [s / mm ² ] is a gradient magnetic field factor representing the degree of signal attenuation due to diffusion, and S ₀ is a signal intensity when the gradient magnetic field factor b is zero.

And
S ₀ = PD (1−exp (−TR / T1) * exp (−TE / T2))
age,
Experiment 1: S ₁ = S ₀ * exp (−b ₁ D)
Experiment 2: S ₂ = S ₀ * exp (−b ₂ D)
Then, the following equation (2) is established.

S ₂ / S ₁ = exp (− (b ₂ −b ₁ ) D) (2)

  Moreover, from the formula (2), the following formula (3) is obtained.

D = −ln (S ₂ / S ₁ ) / (b ₂ −b ₁ ) (3)

  As can be seen from this equation (3), the diffusion coefficient D can be calculated by using two different b values. For this reason, in a general clinical application of DWI, as a simple method, a unidirectional MPG pulse is applied, and diagnosis is performed using a DWI imaged at about b = 1000 and an image imaged at b = 0. Often. In general, since the imaging condition is set so that TE> 60 ms, the image of b = 0 is a T2-weighted image having a contrast that emphasizes the difference in T2.

Here, the b value will be described in detail. In order to understand the b value, first, the gradient magnetic field pulse will be described. In the MRI apparatus, a strong gradient magnetic field that changes linearly along the bore is applied. This gradient magnetic field is called B ₀ field. An MRI apparatus generally includes gradient magnetic field units in the X, Y, and Z directions, and a magnetic field along an arbitrary direction can be applied by combining the gradient magnetic field units. The relationship between the frequency ω of the MR signal and the magnetic field strength B ₀ is expressed by a simple relational expression as shown in Expression (4) below.

ω = γB ₀ (4)

  Here, γ is a proportional constant specific to the nucleus.

  In DWI, spread data is determined by the magnitude and direction of an MPG pulse used for spread encoding. The magnitude of this MPG pulse is the b value. For example, in diffusion encoding, two MPG pulses are used that are arranged in contrast to before and after a high-frequency 180 ° refocus pulse. The first MPG pulse placed before the 180 ° refocus pulse causes a phase shift for all spins.

  Also, the second MPG pulse arranged after the 180 ° refocusing pulse inverts the phase shift caused by the first MPG pulse. This cancels out the phase shift for fixed molecules (protons in medical imaging). However, for a molecule that has moved to a position different from that at the time of the action of the first MPG pulse due to Brownian motion, the phase shift is not completely canceled. Therefore, for such molecules, there will remain a residual phase shift that results in signal attenuation. Thus, in diffusion imaging, diffusion encoding is controlled by the b value indicating the magnitude of the MPG pulse and the direction of the gradient magnetic field pulse.

  The image reconstruction unit 75a performs the above-described DWI to generate a DWI image that emphasizes the diffusion effect in which particles such as water molecules are scattered by Brownian motion due to heat.

  The analysis image generation unit 75b generates various analysis images using the reconstructed image generated by the image reconstruction unit 75a. For example, the analysis image generation unit 75b represents the anisotropy of molecular diffusion in the subject by using the reconstructed image generated by the image reconstructing unit 75a and information on the gradient magnetic field corresponding to each reconstructed image. A DTI image is generated. The analysis image generation unit 75b generates an fMRI image using the reconstructed image generated by the image reconstruction unit 75a and the block setting information.

  Here, DTI and fMRI will be described in detail.

  First, DTI will be described. In DTI, six tensor coefficients or tensor parameters, that is, independent elements or components of a 3 × 3 symmetric tensor matrix are calculated for each voxel based on diffusion data. The symmetric tensor matrix is expressed by the following equation (5).

Here, the first subscript (x, y, z) in D _{xx to} D _zz represents the original direction of the cell or tissue. The second subscript (x, y, z) represents the direction of the gradient magnetic field. Further, D _xx , D _yy , and D _zz that are so-called diagonal components are components of diffusion coefficients that are measured in a three-axis apparatus coordinate system that is usually included in an MRI apparatus used for clinical use.

  Typically, tensor parameters are used to calculate a parameter map that is important for diagnosis. For example, the isotropic component of the diffusion tensor or the anisotropic component of the diffusion tensor is shown in the corresponding parameter map. The parameter map here is, for example, an average apparent diffusion coefficient map or an ADC (Apparent Diffusion Coefficient) map, a partial anisotropy map, or an FA (Fractional Anisotropy) map. The ADC map is a map of diffusion coefficients calculated by solving equation (2) for each pixel.

  Here, since there are a large number of diffusion data used in DTI, when a diffusion tensor is calculated for each voxel, an unknown parameter is determined by averaging. As a method for determining the unknown parameter, a method known by multivariate linear regression, for example, a method of forming a pseudo inverse matrix or a method of performing singular value analysis is used.

For example, according to the method described in Basser, PJ, J. Mattiello, and D. LeBihan. “MR diffusion tensor spectroscopy and imaging” Biophys. J., 66, p.259-267, 1994, voxels in DWI images. Each molecular diffusion anisotropy is expressed by, for example, an ellipsoidal model. In this method, six parameters of diffusion parameters are defined by setting the eigenvalues of the three axes of the ellipsoid to be λ ₁ , λ ₂ , and λ ₃ and the eigenvectors to be v ₁ , v ₂ , and v ₃ .

  For example, the analysis image generation unit 75b defines the anisotropy of molecular diffusion for each voxel included in the DWI image using the ellipsoid model. Then, the analysis image generation unit 75b determines the color of the voxel in the DTI image according to the direction of the longest axis in the defined ellipsoid model. For example, the analysis image generation unit 75b sets the X-axis direction in the apparatus coordinate system to red, the Y-axis direction to yellow, and the Z-axis direction to blue, and the X, Y, Z-axis and the longest axis in the ellipsoid model The voxel color is determined according to the positional relationship with the direction. The analysis image generation unit 75b generates a DTI image representing the anisotropy of molecular diffusion in the subject by performing such processing for each voxel included in the DWI image.

  Next, fMRI will be described. In fMRI, an fMRI image is generated using a BOLD (Blood Oxygenation Level Dependent) effect. Here, the BOLD effect is a decrease in which the contrast of the T2-weighted image changes depending on the degree of oxygenation of hemoglobin in the blood.

  Specifically, in the brain activation region activated by exercise or stimulation, the blood flow increases and oxygen is supplied from the capillaries in the activation region to the nerve cells. Thereby, in the activation region, hemoglobin (oxygenated hemoglobin) combined with oxygen is reduced to reduced hemoglobin. Further, in the activation region, the degree of increase in the oxygen consumption of nerve cells is low with respect to the increase in blood flow, so that the amount of oxyhemoglobin in venous blood relatively increases. Here, oxyhemoglobin is less magnetized than reduced hemoglobin. Therefore, the magnetic susceptibility decreases in the activation region of the brain, and the signal intensity in the T2-weighted image changes.

  In fMRI, using the BOLD effect, a task for activating the motor cortex, visual cortex, auditory cortex, language cortex, sensory cortex, etc. is repeatedly repeated on the subject across the rest during the rest period. As a result, a magnetic resonance image is generated. And a brain function activation site | part can be specified by comparing the image at the time of a task, and the image at the time of a rest. In addition, by comparing images when tasks with different contents are executed, it is possible to identify areas that are activated in common by each task or areas that are specifically activated by each task Can do.

  The following processes are mentioned as an example of the image analysis for specifying a brain function activation area | region. First, an average image of all images at the time of rest and an average image of all images at the time of task are obtained, t-test is performed for both average images to determine a dominant difference from the difference value and the standard error of both mother means, Are created as fMRI original images. Then, a linear correlation coefficient is calculated, and a correlation coefficient between the pixel value of the fMRI original image and the reference function is obtained, thereby creating a correlation coefficient image as an fMRI image.

  Here, the setting of the execution interval between the rest and the task is designed as a block arranged at predetermined intervals. Specifically, the setting of the execution interval between the rest and the task is a design shape in which fixed blocks indicating rest blocks for rest and task blocks for brain activation are arranged in time series. This is called block setting.

  Such block setting is performed by inputting numerical values such as the number of blocks at rest, the number of blocks at task, and the number of repetitions of rest and tasks, as parameters, by an operator such as a doctor. For example, each parameter is input by using a time phase with TR (Repetition Time) as one unit. The block setting is used at the time of image analysis performed after image collection at the time of rest and task.

  The analysis image generation unit 75b performs the fMRI described above to generate an fMRI image depicting the brain function activation region.

  As shown in FIG. 3, the image processing unit 76 includes an image selection unit 76a, an image processing calculation unit 76b, an alignment processing unit 76c, and a tractography generation unit 76d.

  The image selection unit 76a selects, as a reference image, an analysis image related to the subject to be clinically selected from the analysis images stored in the analysis image storage unit 74c.

  Specifically, the image selection unit 76a refers to the analysis image stored by the analysis image storage unit 74c and searches for an analysis image related to the subject to be clinically targeted. At this time, when the identification information for identifying the subject is given to each analysis image, the image selection unit 76a searches for the analysis image based on the identification information.

  Then, when a plurality of types of analysis images related to the clinical subject are searched, the image selection unit 76a determines the type of clinical to be performed. At this time, for example, the image selection unit 76a receives an input of information indicating the clinical type from the operator via the input unit 72, and determines the type of clinical to be performed based on the input information. Thereafter, the image selection unit 76a selects an analysis image suitable as a clinical reference image from the searched analysis images according to the determined clinical type, and the selected analysis image is stored in the internal memory as a reference image. save.

  For example, when the influence of the brain tumor is diagnosed, the image selection unit 76a selects the DWI image of the head related to the clinical subject. Further, for example, when an operation plan related to brain surgery is studied, the image selection unit 76a selects a brain fMRI image related to a subject to be clinically treated. In addition, for example, when follow-up observation or prognosis estimation regarding cerebral infarction or stroke is performed, the image selection unit 76a selects a PWI image and a DWI image of the head related to the subject to be clinically tested. When a plurality of observation images of the same type related to the same subject are searched, the image selection unit 76a selects, for example, the observation image with the latest shooting date from the plurality of search images of the same type. To do.

  The image processing calculation unit 76b sets a region of interest on the reference image based on the image information included in the clinical reference image. In the first embodiment, the image processing calculation unit 76b sets a region of interest on the reference image based on the image information included in the reference image selected by the image selection unit 76a.

  Specifically, the image processing calculation unit 76b reads the reference image selected by the image selection unit 76a from the internal memory, and sets a region of interest according to the type of the read reference image. Here, the setting of the region of interest by the image processing calculation unit 76b will be described with reference to FIGS. In this case, the region of interest is set using the DWI image of the head, the region of interest is set using the fMRI image of the brain, and the region of interest is set using the PWI image and the DWI image of the head. The case of setting will be described as an example.

  FIGS. 4-1 to 4-3 are diagrams for explaining setting of a region of interest when a DWI image of the head is used. For example, when the DWI image of the head is selected by the image selection unit 76a, the image processing calculation unit 76b arbitrarily selects the DWI image via the input unit 72 as illustrated in FIG. An operation for setting the seed point S is received from the operator. At this time, for example, the operator sets a seed point S at a location where a brain tumor is recognized on the DWI image.

  When the seed point is set by the operator, the image processing calculation unit 76b extracts the lesion area A by performing region growing on the basis of the set seed point as shown in FIG. To do. Note that the process for extracting the lesion area A is not limited to region growing, and various generally known area extraction techniques can be used.

In addition, after extracting the lesion area A, the image processing calculation unit 76b generates a margin area M in which the extracted lesion area A is expanded by a predetermined size. Further, the image processing calculation unit 76b divides the area outside the lesion area A into a plurality of divided margin area M _d in the margin region M. At this time, the image processing calculation unit 76b divides the region in the front-rear direction of the head, for example, at a predetermined interval.

Thereafter, the image processing calculation portion 76b, as shown in Figure 4-3, select any one of the region from the plurality of divided margin area M _d, it sets the selected region as a region of interest R _o. In this case, the image processing calculation section 76b via the input section 72 accepts an operation of selecting a region arbitrarily from among a plurality of divided margin area M _d from the operator. Then, the image processing calculation unit 76b sets the region selected by the operator as the region of interest _Ro . Alternatively, the image processing calculation unit 76b, without accepting the operation by the operator, selects the closest area from the seed point S of the plurality of divided margin area M _d, and set the selected region as a region of interest R _o Also good.

As described above, in the first embodiment, the image processing calculation unit 76b divides a region outside the lesion region A in the margin region M , and sets one of the divided regions as a region of interest. This makes it easier for the operator to observe the nerve bundle being pressed by the brain tumor when the tractography generation unit 76d described below generates a curve indicating the nerve bundle using the region of interest as the calculation start region.

The image processing calculation unit 76b is in after setting a region of interest R _o, on the basis of the shape of the head which is depicted in DWI image, calculates a center line which divides the head into right and left. Then, the image processing calculation unit 76b is a symmetrical position calculated centerline as an axis, to set the region of interest R _m having a symmetric shape in the lateral region of interest R _o.

Further, the image processing calculation section 76b via the input section 72 accepts an operation for each region of interest R _o and R _m was set from the operator. Then, the image processing calculation unit 76b, in accordance with the accepted operation to change the shape and position of the region of interest R _o and R _m.

  As described above, in the first embodiment, the image processing calculation unit 76b sets the regions of interest at symmetrical positions about the center line that divides the head into left and right. Thus, when the tractography generation unit 76d described later generates a curve indicating the nerve bundle with the region of interest as the calculation start region, the operator compares the normal nerve bundle with the nerve bundle affected by the affected part. While observing.

FIG. 5 is a diagram for explaining setting of a region of interest when an fMRI image is used. For example, the image processing calculation unit 76b, when fMRI image of the brain is selected by the image selecting section 76a set, as shown in FIG. 5, a brain function activation region B which is depicted in fMRI as a region of interest R _o To do. The image processing calculation unit 76b as the axial center line which divides the head into right and left, the position and shape as the region of interest R _o to set a symmetric region of interest R _m to the left and right.

  FIG. 6 is a diagram for explaining setting of a region of interest using a PWI image and a DWI image of the head. In the examination of cerebral infarction, the usefulness of evaluation combining a DWI image and a PWI image taken by an MRI apparatus is recognized. Specifically, in the examination of cerebral infarction using a DWI image and a PWI image, a segmentation region including a portion suspected of cerebral infarction is extracted from each of the DWI image and the PWI image, and the extracted segmentation regions are overlapped. A region that sometimes does not match (hereinafter referred to as a mismatch area) is identified. The region specified here is called a reversible ischemic region (penumbra region), and is regarded as a region that can be rescued by early resumption of blood flow. Therefore, accurately identifying this reversible ischemic region is very important for diagnosis and treatment of cerebral infarction.

Therefore, for example, when the PWI image and the DWI image of the head are selected by the image selection unit 76a, the image processing calculation unit 76b selects the segmentation region including the part suspected of cerebral infarction, respectively. Extract from For example, as illustrated in FIG. 6, the image processing calculation unit 76 b extracts a segmentation region I _d from the DWI image and extracts a segmentation region I _p from the PWI image.

Thereafter, the image processing calculation unit 76b aligns the positions of the segmentation areas based on the position information attached to the images, and then superimposes the segmentation areas. Then, the image processing calculation unit 76b specifies a region where the segmentation region I _d and the segmentation region I _p do not overlap as a mismatch area, and sets the specified mismatch region as the region of interest R.

  Returning to FIG. 3, the alignment processing unit 76 c aligns the DTI image generated by the analysis image generating unit 75 b and the reference image. In the present embodiment, the alignment processing unit 76c aligns the reference image selected by the image selection unit 76a and the DTI image.

  Specifically, the alignment processing unit 76c reads out the DTI image generated by the analysis image generation unit 75b from the analysis image storage unit 74c. The alignment processing unit 76c reads the reference image selected by the image selection unit 76a from the internal memory. Then, the alignment processing unit 76c aligns the DTI image and the reference image.

  At this time, the alignment processing unit 76 c receives an operation for setting a landmark for each of the DTI image and the reference image from the operator via the input unit 72. The alignment processing unit 76c aligns the DTI image and the reference image by aligning the positions of the landmarks set by the operator.

  Alternatively, the image processing calculation unit 76b may automatically perform alignment between the DTI image and the reference image by using various commonly known image alignment techniques without receiving an operation by the operator. . Examples of the image alignment technique here include “Markerless three-dimensional alignment of CT-MR images” by Yuko Tanaka et al., Medical Electronics and Biotechnology, Vol 34, No. 3, p. 56-65, 1996 can be used.

  The tractography generation unit 76d sets a calculation start region on the DTI image aligned with the reference image by the alignment processing unit 76c based on the information on the region of interest set on the reference image by the image processing calculation unit 76b. Then, a DTT image is generated based on the calculation start area.

  FIG. 7 is a diagram for explaining the generation of the DTT image by the tractography generation unit 76d according to the first embodiment. FIG. 7 shows a part of the DTI image registered with the reference image by the alignment processing unit 76c. Here, for convenience of explanation, a case where a two-dimensional DTI image is used will be described.

  In FIG. 7, a plurality of square areas divided in a lattice form indicate pixels included in the DTI image. Further, the ellipse in each pixel represents an ellipsoidal model representing the molecular diffusion anisotropy described above. Further, the color assigned to each pixel is assigned according to the magnitude of anisotropy, and the darker the color, the lower the anisotropy.

  As shown in FIG. 7, the tractography generation unit 76d first sets the region of interest set on the reference image by the image processing calculation unit 76b as the calculation start region C on the DTI image. Further, the tractography generation unit 76d calculates an average direction for pixels having high anisotropy (white pixels shown in FIG. 7) among pixels included in the DTI image.

Thereafter, the tractography generation unit 76d generates curves F ₁ and F ₂ along the calculated average direction with the pixels P ₁ and P ₂ included in the calculation start region C as base points. The curves F ₁ and F ₂ generated here are trajectories in the maximum diffusion direction in each pixel, and represent the flow of the nerve bundle. In this way, the tractography generation unit 76d generates an image showing the flow of the nerve bundle as a DTT image.

  Returning to the description of FIG. 3, the control unit 77 includes a display control unit 77a. The display control unit 77 a controls display of various types of information on the display unit 73. For example, the display control unit 77a causes the display unit 73 to display the DTT image generated by the tractography generation unit 76d.

8 and 9 are diagrams illustrating examples of DTT images displayed by the display control unit 77a according to the first embodiment. For example, as illustrated in FIG. 8, the display control unit 77 a displays a two-dimensional DTT image superimposed on the two-dimensional DWI image. In FIG. 8, a curve F _o is generated by DTT based on the calculation start area C _o , and the curve F _m is DTT based on a calculation start area C _m symmetrical to the calculation start area C _o. Is generated. By displaying the DTT curves F _o and F _m in this way, the state of the nerve bundle can be observed in a plane.

Alternatively, for example, as illustrated in FIG. 9, the display control unit 77 a combines and displays a three-dimensional DTT image with a two-dimensional DWI image. 9, curve F _o, based on the calculation start region C _o has been generated by DTT, curve F _m, based on the symmetrical calculation start region C _m to the left and right and the calculation start region C _o DTT Is generated. By displaying the DTT curves F _o and F _m in this way, the state of the nerve bundle can be observed in three dimensions.

  Next, a flow of creating a DTT image by the MRI apparatus 100 according to the first embodiment will be described. FIG. 10 is a flowchart illustrating a flow of creating a DTT image by the MRI apparatus 100 according to the first embodiment. As illustrated in FIG. 10, in the MRI apparatus 100 according to the first embodiment, first, the control unit 77 controls each unit such as the sequence control unit 50 based on the imaging conditions set by the operator, so that the expansion is performed. Imaging is performed (step S101).

  Subsequently, the analysis image generation unit 75b generates a DTI image representing the anisotropy of molecular diffusion in the subject (step S102). Further, the image selection unit 76a selects the reference image of the subject from the medical images stored by the reconstructed image storage unit 74b (Step S103).

  Thereafter, the image processing calculation unit 76b sets a region of interest on the reference image based on the image information included in the reference image selected by the image selection unit 76a (step S104). The alignment processing unit 76c aligns the DTI image generated by the analysis image generating unit 75b and the reference image selected by the image selecting unit 76a (step S105).

  Subsequently, the tractography generation unit 76d sets the region of interest set on the reference image by the image processing calculation unit 76b as a calculation start region on the DTI image aligned with the reference image by the alignment unit (step S106). ). Thereafter, the tractography generation unit 76d generates a DTT image based on the set calculation start region (step S107).

  Then, the display control unit 77a displays the DTT image generated by the tractography generation unit 76d on the display unit 73 (step S108).

  As described above, in the first embodiment, the analysis image generation unit 75b generates a DTI image representing the anisotropy of molecular diffusion in the subject. Further, the image processing calculation unit 76b sets a region of interest on the reference image based on the image information included in the clinical reference image. Further, the alignment processing unit 76c aligns the DTI image generated by the analysis image generating unit 75b and the reference image. Then, the tractography generation unit 76d calculates the calculation start region on the DTI image aligned with the reference image by the alignment processing unit 76c based on the information on the region of interest set on the reference image by the image processing calculation unit 76b. And a DTT image is generated based on the calculation start area. Therefore, according to the first embodiment, since a DTT image is generated based on a reference image suitable for clinical use, a DTT image useful in clinical practice can be generated.

In the first embodiment, the image processing calculation unit 76b is, after extracting the lesion area on the reference image, to generate a margin region expanded the lesion area by a predetermined size, the lesion in the margin area A region outside the region is set as a region of interest. Therefore, according to the first embodiment, it is possible to generate a DTT image suitable for diagnosing the influence of a lesion site.

  In the first embodiment, when an fMRI image is used as a reference image, the image processing calculation unit 76b sets a brain function activation region drawn on the fMRI as a region of interest. Therefore, according to the first embodiment, it is possible to generate a DTT image suitable for examining an operation plan related to brain surgery. In addition, this makes it possible for the surgeon to perform an operation in consideration of the local brain function after the operator considers the dominant brain.

  Further, in the first embodiment, when the brain DWI image and PWI image are used as the reference image, the image processing calculation unit 76b extracts a region including a portion suspected of having a cerebral infarction from each of the DWI image and the PWI image. A region where the extracted regions do not overlap is set as a region of interest. Therefore, according to the first embodiment, it is possible to generate a DTT image suitable for follow-up observation and prognosis estimation regarding cerebral infarction or stroke. This also makes it possible for an operator such as a doctor or engineer to easily observe the white matter running state when a brain tumor is present.

  In the first embodiment, the analysis image storage unit 74c stores various analysis images. Further, the image selection unit 76a selects, as a reference image, an analysis image related to the subject to be clinically selected from the analysis images stored by the analysis image storage unit 74c. Then, the image processing calculation unit 76b sets a region of interest on the reference image based on the image information included in the reference image selected by the image selection unit 76a. The alignment processing unit 76c aligns the reference image selected by the image selection unit 76a and the diffusion tensor image. Therefore, according to the first embodiment, even when a plurality of medical images are captured by the MRI apparatus 100, a reference image of a subject to be clinically selected is automatically selected. Images can be easily generated.

  In the first embodiment, the tractography generation unit 76d sets a region of interest set on the reference image as a calculation start region on the DTI image, and generates a DTT image based on the calculation start region. explained. However, the present invention is not limited to this. For example, the region of interest set on the reference image may be presented to the operator once before the DTT image is generated, and changes to the region of interest may be accepted.

  Hereinafter, such a case will be described as a second embodiment. The MRI apparatus according to the second embodiment basically has the same configuration as that of the MRI apparatus 100 illustrated in FIG. 1, and only the processes performed by the image processing unit and the control unit are different. Thus, the second embodiment will be described focusing on processing related to the image processing unit and the control unit.

  FIG. 11 is a functional block diagram illustrating configurations of the storage unit 74, the image processing unit 86, and the control unit 87 according to the second embodiment. In addition, about the function part which plays the role similar to each part shown in FIG. 3, detailed description is abbreviate | omitted as attaching | subjecting the same code | symbol here.

  As shown in FIG. 11, the control unit 87 includes a display control unit 87a and an operation receiving unit 87b.

  The display control unit 87 a controls display of various types of information on the display unit 73. In the second embodiment, the display control unit 87a sets the region of interest set on the reference image by the image processing calculation unit 76b on the DTI image registered by the registration processing unit 76c, and the interest. The area is displayed on the display unit 73. At this time, the display control unit 87 a also causes the display unit 73 to display a GUI for changing or approving the region of interest displayed on the display unit 73.

  The operation receiving unit 87b receives an operation for changing the region of interest displayed on the display unit 73 by the display control unit 87a from the operator. Specifically, the operation accepting unit 87b accepts an operation for the GUI displayed by the display control unit 87a via the input unit 72. When an operation for changing or approving the region of interest is received, the operation receiving unit 87b sends information indicating the content of the operation to the tractography generating unit 86d.

  As shown in FIG. 11, the image processing unit 86 includes an image selection unit 76a, an image processing calculation unit 76b, an alignment processing unit 76c, and a tractography generation unit 86d.

  The tractography generation unit 86d sets a calculation start region on the DTI that is aligned with the reference image by the alignment processing unit 76c based on the information regarding the region of interest that has been changed by the operation reception by the operation reception unit 87b. A DTT image is generated based on the calculation start area.

  Specifically, when information indicating that the region of interest has been changed or approved is sent from the operation receiving unit 87b, the tractography generating unit 86d adjusts the position and size of the region of interest based on the information. Then, the tractography generation unit 86d sets the adjusted region of interest as a calculation start region, and generates a DTT image in the same manner as the method described in the first embodiment.

  Next, a flow of creating a DTT image by the MRI apparatus according to the second embodiment will be described. FIG. 12 is a flowchart illustrating a flow of creating a DTT image by the MRI apparatus according to the second embodiment. The processes in steps S201 to S205 shown in FIG. 12 are the same as the processes in steps S101 to S105 shown in FIG.

  As shown in FIG. 12, in the MRI apparatus according to the second embodiment, the display control unit 87a sets the region of interest set on the reference image by the image processing calculation unit 76b on the DTI image (step S206). Thereafter, the display control unit 87a displays the region of interest set on the DTI image on the display unit 73 (step S207).

  Subsequently, when the operation receiving unit 87b receives an operation for changing or approving the region of interest (Yes in step S208), the tractography generating unit 86d sets the region of interest changed or approved by the operator as the calculation start region. Setting is made (step S209). Thereafter, the tractography generation unit 86d generates a DTT image based on the set calculation start region (step S210).

  Then, the display control unit 87a displays the DTT image generated by the tractography generation unit 76d on the display unit 73 (step S211).

  As described above, in the second embodiment, the analysis image generation unit 75b generates a DTI image representing the molecular diffusion anisotropy in the subject. Further, the image processing calculation unit 76b sets a region of interest on the reference image based on the image information included in the clinical reference image. Further, the alignment processing unit 76c aligns the DTI image generated by the analysis image generating unit 75b and the reference image. Further, the display control unit 87a sets the region of interest set on the reference image by the image processing calculation unit 76b on the DTI image registered by the registration processing unit 76c, and displays the region of interest on the display unit. 73 is displayed. In addition, the operation receiving unit 87b receives an operation for changing the region of interest displayed on the display unit 73 by the display control unit 87a from the operator. Then, the tractography generation unit 86d sets a calculation start region on the DTI image aligned with the reference image by the alignment processing unit 76c based on the information regarding the region of interest changed by the operation reception by the operation reception unit 87b. The DTT image is generated based on the calculation start area. Therefore, according to the second embodiment, the region of interest set on the reference image can be presented to the operator once before the DTT image is generated, and changes to the region of interest can be accepted. A more accurate DTT image can be generated.

  In the first and second embodiments, the case where the MR image generated by the MRI apparatus is used as the reference image has been described. However, the present invention is not limited to this. For example, the present invention can be similarly applied even when other types of images such as a CT image generated by an X-ray CT apparatus and a PET image generated by a PET apparatus are used as a reference image.

DESCRIPTION OF SYMBOLS 100 MRI apparatus 70 Computer system 73 Display part 74 Storage part 74a Raw data storage part 74b Reconstructed image storage part 74c Analysis image storage part 75 Data processing part 75a Image reconstruction part 75b Analysis image generation part 76,86 Image processing part 76a Image Selection unit 76b Image processing calculation unit 76c Registration processing unit 76d, 86d Tractography generation unit 77, 87 Control unit 77a, 87a Display control unit 87b Operation reception unit

Claims (9)

After extracting a lesion area on a clinical reference image , a margin area is generated by expanding the lesion area by a predetermined size, and an area outside the lesion area in the margin area is set as a region of interest. Region of interest setting means to be set;
Alignment means for aligning a diffusion tensor image representing anisotropy of molecular diffusion in the subject and the reference image;
An image for setting a calculation start region on a diffusion tensor image aligned with the reference image based on information on the region of interest set on the reference image, and generating a diffusion tensor tract image based on the calculation start region An image processing apparatus comprising: a generating unit. When a brain DWI image and PWI image are used as clinical reference images, a region including a portion suspected of having a cerebral infarction is extracted from each of the DWI image and the PWI image, and the extracted regions do not overlap each other A region of interest setting means for setting as a region of interest;
Alignment means for aligning a diffusion tensor image representing anisotropy of molecular diffusion in the subject and the reference image;
An image for setting a calculation start region on a diffusion tensor image aligned with the reference image based on information on the region of interest set on the reference image, and generating a diffusion tensor tract image based on the calculation start region Generation means and
An image processing apparatus comprising: After extracting a lesion area on a clinical reference image , a margin area is generated by expanding the lesion area by a predetermined size, and an area outside the lesion area in the margin area is set as a region of interest. Region of interest setting means to be set;
Alignment means for aligning a diffusion tensor image representing anisotropy of molecular diffusion in the subject and the reference image;
Display control means for setting a region of interest set on the reference image on a diffusion tensor image aligned with the reference image, and displaying the region of interest on a display unit;
Operation accepting means for accepting an operation for changing the region of interest displayed on the display unit from an operator;
An image for setting a calculation start region on a diffusion tensor image aligned with the reference image based on information on the region of interest changed by receiving the operation, and generating a diffusion tensor tract image based on the calculation start region An image processing apparatus comprising: a generating unit. When a brain DWI image and PWI image are used as clinical reference images, a region including a portion suspected of having a cerebral infarction is extracted from each of the DWI image and the PWI image, and the extracted regions do not overlap each other A region of interest setting means for setting as a region of interest;
Alignment means for aligning a diffusion tensor image representing anisotropy of molecular diffusion in the subject and the reference image;
Display control means for setting a region of interest set on the reference image on a diffusion tensor image aligned with the reference image, and displaying the region of interest on a display unit;
Operation accepting means for accepting an operation for changing the region of interest displayed on the display unit from an operator;
An image for setting a calculation start region on a diffusion tensor image aligned with the reference image based on information on the region of interest changed by receiving the operation, and generating a diffusion tensor tract image based on the calculation start region Generation means and
An image processing apparatus comprising: Image storage means for storing medical images;
Image selection means for selecting a medical image relating to a clinical subject from the stored medical images as the reference image;
The region-of-interest setting means sets a region of interest on the reference image based on image information included in the selected reference image,
It said alignment means, an image processing apparatus according to any one of claims 1-4, characterized in that for aligning the diffusion tensor image and the reference image selected. First image generation means for generating a diffusion tensor image representing anisotropy of molecular diffusion in a subject from data collected using a magnetic resonance phenomenon;
After extracting a lesion area on a clinical reference image , a margin area is generated by expanding the lesion area by a predetermined size, and an area outside the lesion area in the margin area is set as a region of interest. Region of interest setting means to be set;
Alignment means for aligning the diffusion tensor image and the reference image;
A calculation start region is set on the diffusion tensor image aligned with the reference image based on information on the region of interest set on the reference image, and a diffusion tensor tract image is generated based on the calculation start region. And a magnetic resonance imaging apparatus. First image generation means for generating a diffusion tensor image representing anisotropy of molecular diffusion in a subject from data collected using a magnetic resonance phenomenon;
When a brain DWI image and PWI image are used as clinical reference images, a region including a portion suspected of having a cerebral infarction is extracted from each of the DWI image and the PWI image, and the extracted regions do not overlap each other A region of interest setting means for setting as a region of interest;
Alignment means for aligning the diffusion tensor image and the reference image;
A calculation start region is set on the diffusion tensor image aligned with the reference image based on information on the region of interest set on the reference image, and a diffusion tensor tract image is generated based on the calculation start region. 2 image generation means and
A magnetic resonance imaging apparatus comprising: First image generation means for generating a diffusion tensor image representing anisotropy of molecular diffusion in a subject from data collected using a magnetic resonance phenomenon;
After extracting a lesion area on a clinical reference image , a margin area is generated by expanding the lesion area by a predetermined size, and an area outside the lesion area in the margin area is set as a region of interest. Region of interest setting means to be set;
Alignment means for aligning the diffusion tensor image and the reference image;
Display control means for setting a region of interest set on the reference image on a diffusion tensor image aligned with the reference image, and displaying the region of interest on a display unit;
Operation accepting means for accepting an operation for changing the region of interest displayed on the display unit from an operator;
A calculation start region is set on the diffusion tensor image aligned with the reference image based on the information on the region of interest changed by receiving the operation, and a diffusion tensor tract image is generated based on the calculation start region. And a magnetic resonance imaging apparatus. First image generation means for generating a diffusion tensor image representing anisotropy of molecular diffusion in a subject from data collected using a magnetic resonance phenomenon;
When a brain DWI image and PWI image are used as clinical reference images, a region including a portion suspected of having a cerebral infarction is extracted from each of the DWI image and the PWI image, and the extracted regions do not overlap each other A region of interest setting means for setting as a region of interest;
Alignment means for aligning the diffusion tensor image and the reference image;
Display control means for setting a region of interest set on the reference image on a diffusion tensor image aligned with the reference image, and displaying the region of interest on a display unit;
Operation accepting means for accepting an operation for changing the region of interest displayed on the display unit from an operator;
A calculation start region is set on the diffusion tensor image aligned with the reference image based on the information on the region of interest changed by receiving the operation, and a diffusion tensor tract image is generated based on the calculation start region. 2 image generation means and
A magnetic resonance imaging apparatus comprising:

Images