JP3490161B2 - Method for processing anisotropic flow information image by MR angiography and MRI apparatus - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an anisotropic flow information image processing method, a flow information MR imaging method, and an MRI apparatus. More specifically, the present invention relates to an anisotropic flow information image processing method, a flow information MR imaging method, and an MRI apparatus capable of reducing the difference in blood vessel extraction ability depending on the flow direction.

[0002]

2. Description of the Related Art FIG. 25 shows a two-dimensional P in an MRI apparatus.
C Phase Contrast Angiography
2 is an example of the pulse sequence of FIG. A1, A2, A3
Are a first axis, a second axis and a third axis which are orthogonal to each other in the real space. The first axis A1, the second axis A2, and the third axis A3 are dependent on the slice position. That is, the first axis A1 and the second axis A2 are axes parallel to the slice plane, and the third axis is an axis orthogonal to the slice plane. On the other hand, the X-axis, Y-axis, and Z-axis of the gradient magnetic field coil described later are independent of the slice position. When the first axis A1, the second axis A2, and the third axis A3 are aligned with the three axes X, Y, and Z of the gradient magnetic field coil, the slice plane is orthogonal to the Z axis. This pulse sequence PS
In 1, the RF pulse Rα is applied and the slice gradient s3 is applied to the third axis A3 at timing t0 to excite protons in the slice region. Next, at timing t1,
A flow encode pulse f1 + which is a bipolar pulse whose polarity is inverted from positive to negative is applied to the first axis A1. Next, at timing t2, the warp gradient w1 is applied to the first axis A1 to perform phase encoding. Next, at a timing t4, the read gradient r2 is applied to the second axis A2 to perform frequency encoding, and the echo E is read in the number of data in the read direction (for example, 2).
56) is sampled and the data for the number of data in the read direction is collected. This is repeated only for the number of data in the warp direction (for example, 128) while changing the warp gradient w1, and the data for the number of data for the read direction (for example, 256) × the number of data for the warp direction (for example, 128) is collected. Then, the first axis forward direction image is generated from the data. Re in the figure is a rephase gradient. Further, de is a dephase gradient. Next, in place of the flow encode pulse f1 +, a flow encode pulse f1- that is a bipolar pulse whose polarity is inverted from negative to positive is used.
56) × the number of data in the warp direction (for example, 128) is collected. Then, the first axis negative direction image is generated from the data. Next, the difference between the first-axis positive direction image and the first-axis negative direction image is calculated to obtain the first-axis flow information image (see FIG. 2).
7) is generated.

Next, at timing t5, the RF pulse Rα
And a slice gradient s3 is applied to the third axis A3,
Excite the protons in the slice region. Next, at timing t6, a flow encode pulse f2 +, which is a bipolar pulse whose polarity is inverted from positive to negative, is applied to the second axis A2. Next, at timing t7, the warp gradient w1 is applied to the first axis A1 to perform phase encoding. Next, at timing t9, the read gradient r2 is applied to the second axis A2 to perform frequency encoding, and the echo E is sampled by the number of read direction data (for example, 256) to collect data for the number of read direction data. . This is repeated only for the number of data in the warp direction (for example, 128) while changing the warp gradient w1, and the data for the number of data for the read direction (for example, 256) × the number of data for the warp direction (for example, 128) is collected. Then, a second axis forward direction image is generated from the data. Next, instead of the flow encode pulse f2 +, a flow encode pulse f2- that is a bipolar pulse whose polarity is inverted from negative to positive is used, and otherwise the same as above, the number of read direction data (for example, 256) × warp Data for the number of direction data (for example, 128) is collected.
Then, the second axis negative direction image is generated from the data.
Next, the difference between the second-axis positive direction image and the second-axis negative direction image is calculated to generate the second-axis flow information image (see FIG. 28).

Finally, the first axis flow information image (see FIG. 27).
(Refer to FIG. 28) and the second axis flow information image (refer to FIG. 28) are combined to generate a combined flow information image (refer to FIG. 29).

FIG. 26 shows the flow in the direction of the first axis A1 and the second flow.
It shows a subject with a flow in the direction of the axis A2. When the pulse sequence can PS1 is applied to this subject, the first axis flow information image as shown in FIG. 27 (the flow in the first axis A1 direction is visualized) and the second axis as shown in FIG. A flow information image (a flow in the second axis A2 direction is drawn) is obtained. Then, the first axis flow information image and the second axis flow information image
By synthesizing the axial flow information images, a synthetic flow information image (a flow in the first axis A1 direction and a flow in the second axis A2 direction is depicted) as shown in FIG. 29 is obtained.

FIG. 30 is an example of a pulse sequence for three-dimensional PC angiography in a conventional MRI apparatus. In this pulse sequence PS54, the RF pulse Rα is applied and the slice gradient s3 is applied at the timing t0.
Is applied to the third axis A3 to excite protons in the slice region. Next, at a timing t1, a flow encode pulse f1 + of a bipolar pulse whose polarity is inverted from positive to negative.
Is applied to the first axis A1. Next, at timing t2, the warp gradient w1 is applied to the first axis A1 to perform phase encoding. At the same time, the depth gradient d3 is applied to the third axis A3 to perform phase encoding. Next, at a timing t4, the read gradient r2 is applied to the second axis A2 to perform frequency encoding, and the echo E is read in the number of data in the read direction (for example, 25).
6) Only sample and collect data for the number of data in the read direction. While changing the warp gradient w1 and the depth gradient d3, the number of data in the warp direction (for example, 128)
The number of times of data in the depth direction (for example, 64) is repeated to collect data for the number of data for the read direction (for example, 256) × the number of data for the warp direction (for example, 128) × the number of data for the depth direction (for example, 64). Then, the first-axis positive direction images for the number of depth direction data are generated from the data. Next, instead of the flow encode pulse f1 +, a flow encode pulse f1- that is a bipolar pulse whose polarity is inverted from negative to positive is used, and otherwise the same as above, the number of read direction data (for example, 256) × warp Data of the number of direction data (for example, 128) × the number of depth direction data (for example, 64) is collected. Then, the first axis negative direction images corresponding to the number of depth direction data are generated from the data. Next, the difference between the first axis positive direction image and the first axis negative direction image corresponding to the depth direction is calculated, and the first axis flow information images for the number of depth direction data are generated. Figure 31
Represents a collection of voxels forming these first axis flow information images.

Returning to FIG. 30, at timing t5, the RF pulse Rα is applied and the slice gradient s3 is applied to the third axis A3 to excite protons in the slice region. next,
At timing t6, a flow encode pulse f2 +, which is a bipolar pulse whose polarity is inverted from positive to negative, is applied to the second axis A2. Next, at timing t7, the warp gradient w1 is applied to the first axis A1 to perform phase encoding. at the same time,
The depth gradient d3 is applied to the third axis A3 to perform phase encoding. Next, at timing t9, the read gradient r2 is applied to the second axis A2 to perform frequency encoding, and the echo E is sampled by the number of read direction data (for example, 256) to collect data for the number of read direction data. .
This is repeated by the number of data in the warp direction (for example, 128) × the number of data in the depth direction (for example, 64) while changing the warp gradient w1 and the depth gradient d3, and the number of read direction data (for example, 256) × the number of warp direction data ( For example, 128) × the number of data in the depth direction (for example, 64) is collected. Then, from the data, the second axis positive direction images for the number of depth direction data are generated. Next, instead of the flow encode pulse f2 +, a flow encode pulse f of a bipolar pulse whose polarity is inverted from negative to positive.
2- is used, and otherwise the same as the above, data of the number of read direction data (for example, 256) × the number of warp direction data (for example, 128) × the number of depth direction data (for example, 64) is collected. Then, the second axis negative direction images for the number of depth direction data are generated from the data. Next, the difference between the corresponding second-axis positive direction image and second-axis negative direction image in the depth direction is calculated to generate the second-axis flow information images for the depth direction data number. FIG. 31 shows a group of voxels forming these second axis flow information images.

Returning to FIG. 30, at timing t10, RF
The pulse Rα is applied and the slice gradient s3 is set to the third axis A3.
To excite protons in the slice region. Next, at timing t11, a flow encode pulse f3 +, which is a bipolar pulse whose polarity is inverted from positive to negative, is applied to the third axis A3. Next, at timing t12, the warp gradient w1 is applied to the first axis A1 to perform phase encoding.
At the same time, the depth gradient d3 is applied to the third axis A3 to perform phase encoding. Next, at timing t14, the read gradient r2 is applied to the second axis A2 to perform frequency encoding, and the echo E is read in the read direction data number (for example, 256).
Only the data is sampled and the data for the number of data in the read direction is collected. This is the warp gradient w1 and the depth gradient d3.
The number of data in the warp direction (for example, 128) × the number of data in the depth direction (for example, 64) is repeated while changing Collect (for example, 64) minutes of data. Then, from the data, the third axis positive direction images for the number of depth direction data are generated. Next, in place of the flow encode pulse f3 +, a flow encode pulse f3- that is a bipolar pulse whose polarity is inverted from negative to positive is used, and otherwise the same as above,
Data for the number of data in the read direction (for example, 256) × the number of data for the warp direction (for example, 128) × the number of data for the depth direction (for example, 64) is collected. Then, the third axis negative direction images for the number of depth direction data are generated from the data. Next, the difference between the third axis positive direction image and the third axis negative direction image corresponding to the depth direction is calculated, and the third axis flow information images for the number of depth direction data are generated. FIG. 31 shows a collection of voxels forming these third axis flow information images.

Finally, the corresponding first for depth direction
The axial flow information image, the second axial flow information image, and the third axial flow information image are combined to generate combined flow information images for the number of depth direction data. FIG. 31 shows a collection of voxels that form these combined flow information images.

[0010]

In the pulse sequence can PS1, the number of data in the read direction (for example, 256) is increased in order to improve the resolution, while the number of data in the warp direction (for example, 256) is shortened in order to shorten the scanning time. 12
8) is reduced. For this reason, there is a problem that the resolution in the real space axis direction corresponding to the lead direction and the resolution in the real space axis direction corresponding to the warp direction are different, and the blood vessel extraction capability is different depending on the flow direction. For example, in FIG. 29,
Blood vessels in the first axis A1 direction corresponding to the lead direction are well visualized, but blood vessels in the second axis A2 direction corresponding to the warp direction are difficult to be visualized.

Similarly, the pulse sequence can PS
Also in 54, the number of data in the read direction (for example, 256) is increased in order to increase the resolution, and the number of data in the warp direction (for example, 128) is decreased in order to shorten the scanning time. For this reason, there is a problem that the resolution in the real space axis direction corresponding to the lead direction and the resolution in the real space axis direction corresponding to the warp direction are different, and the blood vessel extraction capability is different depending on the flow direction. Although PC angiography has been described above, TOF angiography (Ti
The same applies to me Of Flight Angiography). Therefore,
An object of the present invention is to provide an anisotropic flow information image processing method, a flow information MR imaging method, and an MRI apparatus capable of reducing the difference in blood vessel extraction ability depending on the flow direction.

[0012]

SUMMARY OF THE INVENTION In a first aspect, the present invention provides an anisotropic flow information image which is composed of anisotropic pixels having different vertical and horizontal sizes and which has flow information as a pixel value. A processing method, characterized by performing an image processing operation for relatively emphasizing the directionality of the anisotropic pixel having a smaller size with respect to the directionality of the anisotropic pixel having a larger size. A method for processing an anisotropic flow image is provided.

According to a second aspect of the present invention, when there is a first axis and a second axis which are orthogonal to each other in real space, a flow encode pulse is added in the direction of the first axis and the number of data in the warp direction is set to be larger than the number of data in the read direction. The data of the anisotropic two-dimensional k-space is reduced to generate the first axis flow information image, the flow encode pulse is added in the second axis direction, and the number of data in the warp direction is smaller than the number of data in the read direction. Then, data of anisotropic two-dimensional k-space is collected to generate a second axis flow information image, and the first axis flow information image and the second axis flow information image are combined to form a combined flow information image. In an MR imaging method of generated flow information, a first axis flow information image is generated with a first axis direction as a warp direction and a second axis direction as a lead direction, and a second axis direction as a lead direction.
An MR imaging method of flow information, characterized in that a second axis flow information image is generated with the axial direction as a warp direction.

According to a third aspect of the present invention, when there is a first axis, a second axis and a third axis which are orthogonal to each other in real space, a flow encode pulse is added in the direction of the first axis and the warp is performed based on the number of data in the read direction. Anisotropic three-dimensional k-space data is collected by reducing the number of direction data to generate first axis flow information images for the number of depth direction data, and a flow encode pulse is added and read in the second axis direction. The number of data in the warp direction is made smaller than the number of data in the direction data to collect anisotropic three-dimensional k-space data, generate the second axis flow information image for the number of data in the depth direction, and encode the flow in the third axis direction. Pulses are added and the number of data in the warp direction is made smaller than the number of data in the read direction to collect anisotropic three-dimensional k-space data to generate the third axis flow information image for the number of data in the depth direction. flow An MR imaging method of flow information for synthesizing a report image, the second axial flow information image, and the third axial flow information image to generate a synthetic flow information image, wherein a first axis direction is a warp direction and a second axis direction is a warp direction. And a third axis is a depth direction to generate a first axis flow information image, a first axis direction is a lead direction, a second axis direction is a warp direction, and a third axis is a depth direction. The first axis direction is the lead direction, the second axis direction is the warp direction, the third axis is the depth direction, and the number of warp direction data is increased to the same as the number of read direction data. An MR imaging method of flow information is provided, characterized in that the third axis flow information image is generated by reducing the number of depth direction data.

According to a fourth aspect, the present invention provides an MRI apparatus for generating an anisotropic flow information image which is composed of anisotropic pixels having different vertical and horizontal sizes and has flow information as pixel values. An image processing operation is performed on the anisotropic flow information image to emphasize the directionality of the anisotropic pixel having a smaller size relative to the directionality of the anisotropic pixel having a larger size. Provided is an MRI apparatus including a calculation means.

In a fifth aspect, the present invention applies a flow encode pulse to either the first axis direction or the second axis direction when there is a first axis and a second axis which are orthogonal to each other in the real space, and the read direction. Data collecting means for collecting anisotropic two-dimensional k-space data by reducing the number of data in the warp direction from the number of data, and a first axis flow information image from the data collected by adding a flow encode pulse in the first axis direction. And a flow information image generating means for generating a second axis flow information image from data collected by adding a flow encode pulse in the second axis direction, the first axis flow information image and the second axis flow information image. And a flow information image synthesizing means for synthesizing the flow information image and a flow information image synthesizing means to generate a synthesized flow information image. And direction, characterized by collecting data and a direction orthogonal thereto as the read direction MR
I device is provided.

According to a sixth aspect of the present invention, when there is a first axis, a second axis and a third axis which are orthogonal to each other in real space, the present invention can be applied to either the first axis direction, the second axis or the third axis direction. Data collecting means for adding anisotropic data to the anisotropic three-dimensional k-space by adding a flow encode pulse, reducing the number of data in the warp direction from the number of data in the read direction and reducing the number of data in the depth direction from the number of data in the warp direction, The number of data in the depth direction is generated from the data collected by adding the flow encode pulse in the second axis direction from the data collected by adding the flow encode pulse in the first axis direction. For generating the second axis flow information image for the minute axis, and for generating the third axis flow information image for the number of data in the depth direction from the collected data by adding the flow encode pulse in the third axis direction. An MRI apparatus having an information image generating means, and a flow information image combining means for combining the first axis flow information image, the second axis flow information image and the third axis flow information image to generate a combined flow information image. In the above, the data collecting means collects data by applying a flow encode pulse in the first axis direction, using the first axis direction as the warp direction, the second axis direction as the lead direction, and the third axis as the depth direction, Data is collected by adding a flow encode pulse to the first direction, setting the first axis direction to the read direction, the second axis direction to the warp direction, and the third axis to the depth direction, adding the flow encode pulse to the third axis direction, and adding the first axis direction. The direction is the lead direction, the second axis is the warp direction, the third axis is the depth direction, and the warp direction data count is increased to the same as the read direction data count. Provides an MRI apparatus wherein collecting data by reducing the number of depth direction data in accordance with the number of data.

[0018]

In the anisotropic flow information image processing method according to the first aspect and the MRI apparatus according to the fourth aspect, the directionality of the anisotropic pixel having a smaller size is compared with the anisotropic pixel size. The image processing operation for emphasizing the larger directionality is performed. Examples of this image processing calculation include an image processing calculation using a line detection template (Template) described later and an image processing calculation using a maximum value filter and a minimum value filter. Since the difference in resolution is leveled by this image processing calculation, it is possible to reduce the difference in blood vessel extraction ability depending on the flow direction.

In the MR imaging method according to the second aspect and the MR imaging apparatus according to the fifth aspect, a flow encode pulse is added to the first axis or the second axis to reduce the number of data in the warp direction from the number of data in the read direction. Data of anisotropic two-dimensional k-space is collected by using the warp direction as the direction to which the flow encode pulse is applied and the read direction as the direction orthogonal thereto. As a result, the resolution in the second axis direction is high during MR imaging of the flow in the first axis direction, and the resolution in the first axis direction is high during MR imaging of flow in the second axis direction. That is, since the resolution of the blood vessel in the radial direction is always high, it is possible to reduce the difference in blood vessel extraction ability depending on the flow direction.

In the MR imaging method according to the third aspect and the MR imaging apparatus according to the sixth aspect, a flow encode pulse is added to the first axis or the second axis to reduce the number of data in the warp direction from the number of data in the read direction. Data of anisotropic three-dimensional k-space is collected by using the direction of applying the flow encode pulse as the warp direction and the direction orthogonal thereto as the read direction. Further, a flow encode pulse is added to the third axis to collect anisotropic three-dimensional k-space data. At that time, the number of warp direction data is increased to the same number as the number of read direction data, and the increased amount is added. The number of data in the depth direction is reduced according to. As a result, M of the flow in the first axial direction
During R imaging, the resolution in the direction of the second axis becomes high,
During MR imaging of the flow in the second axis direction, the resolution in the first axis direction is high. Further, the resolution in the first axis direction and the second axis direction is high during MR imaging of the flow in the third axis direction. That is, since the resolution of the blood vessel in the radial direction is always high, it is possible to reduce the difference in blood vessel extraction ability depending on the flow direction. Further, since the number of data in the depth direction is reduced according to the amount of increase in the number of data in the warp direction during MR imaging of the flow in the third axis direction, it is possible to suppress the extension of the scan time.

[0021]

The present invention will be described in more detail with reference to the embodiments shown in the drawings. The present invention is not limited to this. -First Embodiment- A first embodiment is an embodiment of an anisotropic flow information image processing method of the present invention and an MRI apparatus for implementing the method. FIG. 1 is a block diagram of an MRI apparatus 100 according to an embodiment of the present invention. In this MRI apparatus 100, the magnet assembly 1 has a space portion (hole) for inserting a subject therein, and a main magnetic field for applying a constant main magnetic field to the subject by surrounding this space portion. A coil, a gradient magnetic field coil for generating a gradient magnetic field (the gradient magnetic field coil includes coils of three axes X, Y, and Z orthogonal to each other in real space), and spins of atomic nuclei in the subject. A transmission coil that applies an RF pulse for excitation, a reception coil that detects an NMR (Nuclear Magnetic Resonance) signal from the subject, and the like are arranged. The main magnetic field coil, gradient magnetic field coil, transmitter coil and receiver coil are
They are connected to the main magnetic field power source 2, the gradient magnetic field driving circuit 3, the RF power amplifier 4 and the preamplifier 5, respectively.

The sequence storage circuit 8 operates the gradient magnetic field drive circuit 3 in accordance with a sequence such as the spin-warp method in accordance with a command from the computer 7 to generate a gradient magnetic field from the gradient magnetic field coil of the magnet assembly 1. At the same time, the gate modulation circuit 9 is operated to modulate the high frequency output signal from the RF oscillating circuit 10 into a pulse-shaped signal having a predetermined timing and a predetermined envelope, which is added as an RF pulse to the RF power amplifier 4 and then the RF power amplifier 4 After power amplification by, the signal is applied to the transmission coil of the magnet assembly 1 to selectively excite a desired slice area.

The preamplifier 5 comprises a magnet assembly 1
The NMR signal from the subject detected by the receiving coil is amplified and input to the phase detector 12. The phase detector 12 is
The output of the RF oscillation circuit 10 is used as a reference signal, and the preamplifier 5
The NMR signal from is phase-detected and given to the A / D converter 11. The A / D converter 11 converts the analog signal after phase detection into a digital signal and inputs it to the computer 7.
The computer 7 performs an image reconstruction operation on the digital signal from the A / D converter 11 to generate an image (proton density image) of the slice area. This image is displayed on the display device 6. In addition, the computer 7 is a console 13
Responsible for overall control such as receiving information input from.

The anisotropic flow information image processing method of the present invention is executed under the control of the computer 7. That is, the computer 7 performs an image processing operation for emphasizing the directionality of the anisotropic pixel having a smaller size relative to the directionality of the anisotropic pixel having a larger size on the anisotropic flow information image. It functions as a calculation means to perform.

FIG. 2 shows a pulse sequence for obtaining two-dimensional PC angiography in the MRI apparatus 100. In this pulse sequence PS1, timing t0
Then, the RF pulse Rα is applied and the slice gradient s3 is applied to the third axis A3 to excite the protons in the slice region. Next, at timing t1, a flow encode pulse f1 +, which is a bipolar pulse whose polarity is inverted from positive to negative, is applied to the first axis A1. Next, at timing t2, the warp gradient w1 is applied to the first axis A1 to perform phase encoding. Next, at timing t4, the read gradient r2 is applied to the second axis A2 to perform frequency encoding, and the echo E is sampled by the number of read direction data (for example, 256), and the data of the number of read direction data is collected. . This is repeated only for the number of data in the warp direction (for example, 128) while changing the warp gradient w1, and the number of data for the read direction (for example, 256) × the number of data for the warp direction (for example, 12).
8) Collect minute data. Then, the first axis forward direction image is generated from the data. Re in the figure is a rephase gradient. Further, de is a dephase gradient. Next, in place of the flow encode pulse f1 +, a flow encode pulse f1- that is a bipolar pulse whose polarity is inverted from negative to positive is used.
Data of the number of read direction data (for example, 256) × the number of warp direction data (for example, 128) is collected. Then, the first axis negative direction image is generated from the data. Next, the difference between the first-axis positive direction image and the first-axis negative direction image is calculated to generate the first-axis flow information image (see FIG. 4).

Next, at timing t5, the RF pulse Rα
And a slice gradient s3 is applied to the third axis A3,
Excite the protons in the slice region. Next, at timing t6, a flow encode pulse f2 +, which is a bipolar pulse whose polarity is inverted from positive to negative, is applied to the second axis A2. Next, at timing t7, the warp gradient w1 is applied to the first axis A1 to perform phase encoding. Next, at timing t9, the read gradient r2 is applied to the second axis A2 to perform frequency encoding, and the echo E is sampled by the number of read direction data (for example, 256) to collect data for the number of read direction data. . This is repeated only for the number of data in the warp direction (for example, 128) while changing the warp gradient w1, and the data for the number of data for the read direction (for example, 256) × the number of data for the warp direction (for example, 128) is collected. Then, a second axis forward direction image is generated from the data. Next, instead of the flow encode pulse f2 +, a flow encode pulse f2- that is a bipolar pulse whose polarity is inverted from negative to positive is used, and otherwise the same as above, the number of read direction data (for example, 256) × warp Data for the number of direction data (for example, 128) is collected.
Then, the second axis negative direction image is generated from the data.
Next, the difference between the second-axis positive direction image and the second-axis negative direction image is calculated to generate the second-axis flow information image (see FIG. 5).

Finally, the first axial flow information image (see FIG. 4) and the second axial flow information image (see FIG. 5) are combined to generate a combined flow information image (see FIG. 6). Usually, assuming that the flow image in the x direction is X and the flow image in the y direction is Y, the synthesis operation is generally (X ² + Y ² ) ^1/2, but for simplicity, it is set to (X + Y).

FIG. 3 shows a subject having a flow in the first axis A1 direction and a flow in the second axis A2 direction. When the pulse sequence can PS1 is applied to this subject, the first axis flow information image (first axis A1) as shown in FIG.
Directional flow) and the second as shown in FIG.
Axial flow information image (flow in the second axis A2 direction is depicted)
Is obtained. Then, by combining the first axis flow information image and the second axis flow information image, a combined flow information image as shown in FIG. 6 (a flow in the first axis A1 direction and a flow in the second axis A2 direction is drawn). Is obtained. This synthetic flow information image is
It is an anisotropic flow information image composed of anisotropic pixels having different vertical and horizontal sizes and having flow information as pixel values. In this example, the pixel size in the vertical direction (second axis A2 direction) is small, and the pixel size in the horizontal direction (first axis A1 direction) is large.

Calculator 7 gives the composite flow information image
An image processing calculation is performed to emphasize the directivity in the vertical direction (the one with the smaller anisotropic pixel size) with respect to the directivity in the horizontal direction (the one with the larger anisotropic pixel size). For example, the following image processing calculation is performed using the vertical line detection template g1 shown in FIG. I (i, j) = f (i, j) + α · temp {g1, f (i,
j)} where I (i, j) is the pixel value at the coordinate (i, j) in the processed combined flow information image. i is the coordinate value in the direction of the first axis A1. j is a coordinate value in the second axis A2 direction. Also, f (i, j)
Is the pixel value of the coordinate (i, j) in the combined flow information image before processing. α is an empirically determined coefficient. temp {g1, f
(i, j)} is the next operation. temp {g1, f (i, j)} = f (i, j) + f (i, j-
1) + f (i, j + 1) FIG. 8 shows a combined flow information image as a result of performing the above image processing operation on the combined flow information image of FIG. 6 with α = 1. Comparing FIG. 8 and FIG. 6, it can be seen that in the combined flow information image after processing, the difference in blood vessel extraction ability depending on the flow direction is smaller than in the combined flow information image before processing.

-Modification of the First Embodiment-The computer 7 is anisotropic in the directionality of the smaller anisotropic pixel size with respect to the flow information image for the axis having the smaller anisotropic pixel size. An image processing operation for emphasizing the directionality of the larger size of the sex pixel is performed. For example, the following image processing calculation is performed using the vertical line detection template g1 shown in FIG. E (i, j) = temp {g1, e (i, j)} where E (i, j) is the pixel value at the coordinate (i, j) in the processed flow information image. i is the coordinate value in the direction of the first axis A1.
j is a coordinate value in the second axis A2 direction. e (i, j) is the pixel value of the coordinate (i, j) in the flow information image before processing.
Then, the flow information image for the axis with the larger size of the anisotropic pixels and the flow information image after the above processing are combined,
Obtain a composite flow information image.

FIG. 9 shows a second axis flow information image obtained as a result of performing the above image processing operation on the second axis flow information image of FIG. 5 (an image of a vertical flow having a small anisotropic pixel size). Indicates. Further, FIG. 10 shows a combined flow information image in which the first axis flow information image of FIG. 4 and the second axis flow information image of FIG. 9 are combined. Comparing FIG. 10 and FIG. 6, it can be seen that in the combined flow information image after processing, the difference in blood vessel extraction ability depending on the flow direction is smaller than in the combined flow information image before processing.

-Other Modifications of First Embodiment- A line detection template g2 as shown in FIG. 11 may be used. The following image processing calculation may be performed. I (i, j) = f (i, j) .alpha.temp {g1, f (i,
j)} Alternatively, I (i, j) = α · temp {g1, f (i, j)} Further, the above image processing is performed on the anisotropic flow information image obtained by TOF angiography or three-dimensional angiography. You may perform calculation.

-Second Embodiment- A second embodiment is an embodiment of an MR imaging method (two-dimensional) of flow information of the present invention and an MRI apparatus for implementing the method. The MRI apparatus of the second embodiment is the above-mentioned MRI.
It has the same configuration as the device 100. The flow information MR imaging method (two-dimensional) of the present invention is executed under the control of the computer 7. That is, the computer 7 functions as a data collection unit that collects data with the direction to which the flow encode pulse is added as the warp direction and the direction orthogonal thereto as the read direction.

FIG. 12 shows a pulse sequence for obtaining two-dimensional PC angiography in the MRI apparatus 100. In this pulse sequence PS2, timing t0
Then, the RF pulse Rα is applied and the slice gradient s3 is applied to the third axis A3 to excite the protons in the slice region. Next, at timing t1, a flow encode pulse f1 +, which is a bipolar pulse whose polarity is inverted from positive to negative, is applied to the first axis A1. Next, at timing t2, the warp gradient w1 is applied to the first axis A1 to perform phase encoding. Next, at timing t4, the read gradient r2 is applied to the second axis A2 to perform frequency encoding, and the echo E is sampled by the number of read direction data (for example, 256), and the data of the number of read direction data is collected. . This is repeated only for the number of data in the warp direction (for example, 128) while changing the warp gradient w1, and the number of data for the read direction (for example, 256) × the number of data for the warp direction (for example, 12).
8) Collect minute data. Then, the first axis forward direction image is generated from the data. Next, instead of the flow encode pulse f1 +, a flow encode pulse f1- that is a bipolar pulse whose polarity is inverted from negative to positive is used, and otherwise the same as above, the number of read direction data (for example,
256) × the number of data in the warp direction (for example, 128) is collected. Then, the first axis negative direction image is generated from the data. Next, the difference between the first-axis positive direction image and the first-axis negative direction image is calculated to generate the first-axis flow information image (see FIG. 14).

Next, at timing t5, the RF pulse Rα
And a slice gradient s3 is applied to the third axis A3,
Excite the protons in the slice region. Next, at timing t6, a flow encode pulse f2 +, which is a bipolar pulse whose polarity is inverted from positive to negative, is applied to the second axis A2. Next, at timing t7, the warp gradient w2 is applied to the second axis A2 to perform phase encoding. Next, at timing t9, the read gradient r1 is applied to the first axis A1 to perform frequency encoding, and the echo E is sampled by the number of read direction data (for example, 256) to collect data for the number of read direction data. . This is repeated only for the number of data in the warp direction (for example, 128) while changing the warp gradient w2, and the data for the number of data for the read direction (for example, 256) × the number of data for the warp direction (for example, 128) is collected. Then, a second axis forward direction image is generated from the data. Next, instead of the flow encode pulse f2 +, a flow encode pulse f2- that is a bipolar pulse whose polarity is inverted from negative to positive is used, and otherwise the same as above, the number of read direction data (for example, 256) × warp Data for the number of direction data (for example, 128) is collected.
Then, the second axis negative direction image is generated from the data.
Next, the difference between the second-axis positive direction image and the second-axis negative direction image is calculated to generate the second-axis flow information image (see FIG. 16).

Finally, the first axis flow information image (see FIG. 14).
(See FIG. 16) and the second axis flow information image (see FIG. 16) are combined to generate a combined flow information image (see FIG. 17).

13 and 15 show a subject having a flow in the first axis A1 direction and a flow in the second axis A2 direction.
The pulse sequence sequence PS2 is applied to the subject.
Is applied, a first axis flow information image as shown in FIG. 14 (a flow in the first axis A1 direction is visualized) and a second axis flow information image as shown in FIG. 16 (flow in the second axis A2 direction) are applied. Is drawn). In the first axis flow information image and the second axis flow information image, the resolution in the direction orthogonal to the drawn flow direction (the radial direction of the blood vessel) is high. Then, by combining the first axis flow information image and the second axis flow information image, a combined flow information image as shown in FIG. 17 (flow in the first axis A1 direction and flow in the second axis A2 direction is drawn). Is obtained. This combined flow information image is an isotropic flow information image that is composed of isotropic pixels having the same vertical size and horizontal size and that uses flow information as pixel values. That is, there is no difference in blood vessel extraction ability depending on the flow direction.

The above pulse sequence can PS2
The scan time T required to collect all the data is: T = 2 × (128 + 128) × TR = 512 · TR. However, 2 is a flow encode in the positive direction and the negative direction. 128 + 128 is the number of warp direction data for the first axis A1 and the number of warp direction data for the second axis A2. TR is the repeat time. The scan time T is the same as the scan time required to collect all data by the conventional pulse sequence can PS1.

-Modification of Second Embodiment- FIG. 18 shows another pulse sequence for obtaining two-dimensional PC angiography in the MRI apparatus 100. In this pulse sequence PS3, at timing t0, RF
The pulse Rα is applied and the slice gradient s3 is set to the third axis A3.
To excite protons in the slice region. Next, at timing t1, a flow encode pulse f1 + of a bipolar pulse whose polarity is inverted from positive to negative is applied to the first axis A.
1 is applied. Next, at the timing t2, the warp gradient w
1 is applied to the first axis A1 to perform phase encoding. Next, at timing t4, the read gradient r2 is applied to the second axis A2 to perform frequency encoding, and the echo E is sampled by the read direction data number (for example, 256),
Collect data for the number of data in the read direction. The number of data in the warp direction (for example, 1 while changing the warp gradient w1)
28) times, and the number of data in the read direction (for example,
256) × the number of data in the warp direction (for example, 128) is collected. Then, the first axis forward direction image is generated from the data.

Next, at timing t5, the RF pulse Rα
And a slice gradient s3 is applied to the third axis A3,
Excite the protons in the slice region. Next, at timing t6, a flow encode pulse f2 +, which is a bipolar pulse whose polarity is inverted from positive to negative, is applied to the second axis A2. Next, at timing t7, the warp gradient w2 is applied to the second axis A2 to perform phase encoding. Next, at timing t9, the read gradient r1 is applied to the first axis A1 to perform frequency encoding, and the echo E is sampled by the number of read direction data (for example, 256) to collect data for the number of read direction data. . This is repeated only for the number of data in the warp direction (for example, 128) while changing the warp gradient w2, and the data for the number of data for the read direction (for example, 256) × the number of data for the warp direction (for example, 128) is collected. Then, a second axis forward direction image is generated from the data.

Next, at timing t10, the RF pulse R
α is applied and the slice gradient s3 is applied to the third axis A3 to excite protons in the slice region. Next, without applying the flow encode pulse, at timing t12,
A warp gradient w2 is applied to the second axis A2 to perform phase encoding. Next, at timing t14, the read gradient r1 is applied to the first axis A1 to perform frequency encoding, and the echo E is sampled by the number of read direction data (for example, 256) to collect data for the number of read direction data. . This is repeated only for the number of data in the warp direction (for example, 256) while changing the warp gradient w2, and the data for the number of data for the read direction (for example, 256) × the number of data for the warp direction (for example, 256) is collected. Then, a reference image is generated from the data.

Next, the difference between the first axis forward direction image and the reference image is calculated to generate the first axis flow information image. Also,
The difference between the second axis forward direction image and the reference image is calculated, and the second image
Generate an axial flow information image. Finally, the first axis flow information image and the second axis flow information image are combined to generate a combined flow information image.

This composite flow information image is an isotropic flow information image which is composed of isotropic pixels having the same vertical size and horizontal size and whose flow information is a pixel value. That is, there is no difference in blood vessel extraction ability depending on the flow direction.

The above pulse sequence can PS3
The scan time T required to collect all the data is T = (128 + 128 + 256) × TR = 512 · TR. However, 128 + 128 + 256 is the number of warp direction data for obtaining the first axis flow information image, the number of warp direction data for obtaining the second axis flow information image, and the number of warp direction data for obtaining the reference image. Scan time T
Is the same as the scan time required to collect all the data by the conventional pulse sequence can PS1.

-Third Embodiment- A third embodiment is an embodiment of an MR imaging method (three-dimensional) of flow information of the present invention and an MRI apparatus for implementing the method. The MRI apparatus of the third embodiment has the same configuration as the MRI apparatus 100 of the first embodiment. The flow information MR imaging method (three-dimensional) of the present invention is based on a computer 7
It is executed under the control of. That is, the computer 7
Data is collected by applying a flow encode pulse in the axial direction, using the first axis direction as the warp direction, the second axis direction as the lead direction, and the third axis as the depth direction, adding the flow encode pulse in the second axis direction, and Data is collected with the axial direction as the lead direction, the second axial direction as the warp direction, the third axial direction as the depth direction, and a flow encode pulse is applied to the third axial direction and the first axial direction as the lead direction and the second axial direction as It functions as a data collection means that sets the warp direction, the third axis is the depth direction, increases the number of data in the warp direction to the same number as the number of data in the read direction, and reduces the number of data in the depth direction according to the increased number of data in the warp direction. To do.

FIG. 19 shows a pulse sequence for obtaining three-dimensional PC angiography in the MRI apparatus 100. In this pulse sequence PS4, the timing t0
Then, the RF pulse Rα is applied and the slice gradient s3 is applied to the third axis A3 to excite the protons in the slice region. Next, at timing t1, a flow encode pulse f1 +, which is a bipolar pulse whose polarity is inverted from positive to negative, is applied to the first axis A1. Next, at timing t2, the warp gradient w1 is applied to the first axis A1 to perform phase encoding. At the same time, the depth gradient d3 is applied to the third axis A3 to perform phase encoding. Next, at a timing t4, the read gradient r2 is applied to the second axis A2 to perform frequency encoding, and the echo E is read in the number of data in the read direction (for example, 25).
6) Only sample and collect data for the number of data in the read direction. While changing the warp gradient w1 and the depth gradient d3, the number of data in the warp direction (for example, 128)
The data is collected by the number of data in the depth direction (for example, 64), and the number of data in the read direction (for example, 256) × the number of data in the warp direction (for example, 128) × the number of data in the depth direction (for example, 64). Then, the first-axis positive direction images for the number of depth direction data are generated from the data. Next, instead of the flow encode pulse f1 +, a bipolar pulse flow encode pulse f1- whose polarity is inverted from negative to positive is used, and otherwise the same as above, and the read direction data number (for example, 256) × warp Data of the number of direction data (for example, 128) × the number of depth direction data (for example, 64) is collected. Then, the first axis negative direction images corresponding to the number of depth direction data are generated from the data. Next, the difference between the first axis positive direction image and the first axis negative direction image corresponding to the depth direction is calculated, and the first axis flow information images for the number of depth direction data are generated. Figure 20
Represents a collection of voxels forming these first axis flow information images.

Returning to FIG. 19, at timing t5, the RF pulse Rα is applied and the slice gradient s3 is applied to the third axis A3 to excite protons in the slice region. next,
At timing t6, a flow encode pulse f2 +, which is a bipolar pulse whose polarity is inverted from positive to negative, is applied to the second axis A2. Next, at timing t7, the warp gradient w2 is applied to the second axis A2 to perform phase encoding. at the same time,
The depth gradient d3 is applied to the third axis A3 to perform phase encoding. Next, at timing t9, the read gradient r1 is applied to the first axis A1 to perform frequency encoding, and the echo E is sampled by the number of read direction data (for example, 256) to collect data for the number of read direction data. .
This is repeated for the number of data in the warp direction (for example, 128) × the number of data for the depth direction (for example, 64) while changing the warp gradient w2 and the depth gradient d3, and the number of read direction data (for example, 256) × the number of warp direction data ( For example, 128) × the number of data in the depth direction (for example, 64) is collected. Then, from the data, the second axis positive direction images for the number of depth direction data are generated. Next, instead of the flow encode pulse f2 +, a flow encode pulse f of a bipolar pulse whose polarity is inverted from negative to positive.
2- is used, and otherwise the same as the above, data of the number of read direction data (for example, 256) × the number of warp direction data (for example, 128) × the number of depth direction data (for example, 64) is collected. Then, the second axis negative direction images for the number of depth direction data are generated from the data. Next, the difference between the corresponding second-axis positive direction image and second-axis negative direction image in the depth direction is calculated to generate the second-axis flow information images for the depth direction data number. FIG. 21 shows a group of voxels forming these second axis flow information images.

Returning to FIG. 19, at timing t10, RF
The pulse Rα is applied and the slice gradient s3 is set to the third axis A3.
To excite protons in the slice region. Next, at timing t11, a flow encode pulse f3 +, which is a bipolar pulse whose polarity is inverted from positive to negative, is applied to the third axis A3. Next, at timing t12, the warp gradient w2 ′ is applied to the second axis A2 to perform phase encoding. At the same time, the depth gradient d3 ′ is applied to the third axis A3 to perform phase encoding. Next, at timing t14, the read gradient r1 is applied to the first axis A1 to perform frequency encoding, and the echo E is read in the number of data in the read direction (for example, 2).
56) is sampled and the data for the number of data in the read direction is collected. While changing the warp gradient w2 ′ and the depth gradient d3 ′, the number of data in the warp direction (for example, 2)
56) × depth direction data number (eg, 32) times are repeated to collect data in the read direction data number (eg, 256) × warp direction data number (eg, 256) × depth direction data number (eg, 32) To do. Then, from the data, the third axis positive direction images for the number of depth direction data are generated. Next, instead of the flow encode pulse f3 +, a flow encode pulse f3- which is a bipolar pulse whose polarity is inverted from negative to positive is used, and otherwise the same as above, the number of read direction data (for example, 256) × warp Data of the number of direction data (for example, 256) × the number of depth direction data (for example, 32) is collected. Then, the third axis negative direction images for the number of depth direction data are generated from the data. Next, the difference between the third axis positive direction image and the third axis negative direction image corresponding to the depth direction is calculated, and the third axis flow information images for the number of depth direction data are generated. FIG. 22
Represents a collection of voxels forming these third axis flow information images.

Finally, the corresponding first for depth direction
The axial flow information image, the second axial flow information image, and the third axial flow information image are combined to generate combined flow information images for the number of depth direction data. FIG. 23 shows a collection of voxels that form these combined flow information images.

The above pulse sequence can PS4
, The scan time T required to collect all data is: T = 2 × (128 × 64 + 128 × 64 + 256 × 32) × TR = 49
152.TR. However, 128 x 64 + 128 x 64 + 256 x 32 is the number of warp direction data to obtain the first axis flow information image x depth direction data number and the number of warp direction data to obtain the second axis flow information image x depth direction data number And the number of data in the warp direction for obtaining the third axis flow information image × the number of data in the depth direction. The scan time T is the same as the scan time required to collect all data by the conventional pulse sequence can PS54.

-Modification of Third Embodiment- FIG. 24 shows another pulse sequence for obtaining three-dimensional PC angiography in the MRI apparatus 100. In this pulse sequence PS5, at timing t0, RF
The pulse Rα is applied and the slice gradient s3 is set to the third axis A3.
To excite protons in the slice region. Next, at timing t1, a flow encode pulse f1 + of a bipolar pulse whose polarity is inverted from positive to negative is applied to the first axis A.
1 is applied. Next, at the timing t2, the warp gradient w
1 is applied to the first axis A1 to perform phase encoding. At the same time, the depth gradient d3 is applied to the third axis A3 to perform phase encoding. Next, at a timing t4, the lead gradient r2
Is applied to the second axis A2 to perform frequency encoding, and the echo E is sampled by the number of read direction data (for example, 256) to collect data for the number of read direction data. This is repeated by the number of data in the warp direction (for example, 128) × the number of data in the depth direction (for example, 64) while changing the warp gradient w1 and the depth gradient d3, and the number of read direction data (for example, 256) × the number of warp direction data ( For example, 128) × depth direction data number (for example, 6)
4) Collect minute data. Then, the first-axis positive direction images for the number of depth direction data are generated from the data.

Next, at timing t5, the RF pulse Rα
And a slice gradient s3 is applied to the third axis A3,
Excite the protons in the slice region. Next, at timing t6, a flow encode pulse f2 +, which is a bipolar pulse whose polarity is inverted from positive to negative, is applied to the second axis A2. Next, at timing t7, the warp gradient w2 is applied to the second axis A2 to perform phase encoding. At the same time, the depth gradient d3 is applied to the third axis A3 to perform phase encoding. Next, at timing t9, the read gradient r1 is applied to the first axis A1 to perform frequency encoding, and the echo E is sampled by the number of read direction data (for example, 256) to collect data for the number of read direction data. . This is repeated for the number of data in the warp direction (for example, 128) × the number of data for the depth direction (for example, 64) while changing the warp gradient w2 and the depth gradient d3, and the number of read direction data (for example, 256) × the number of warp direction data ( For example,
128) × the number of data in the depth direction (for example, 64) is collected. Then, from the data, the second axis positive direction images for the number of depth direction data are generated.

Next, at timing t10, the RF pulse R
α is applied and the slice gradient s3 is applied to the third axis A3 to excite protons in the slice region. Next, at timing t11, a flow encode pulse f3 +, which is a bipolar pulse whose polarity is inverted from positive to negative, is applied to the third axis A3. Next, at timing t12, the warp gradient w2 ′
Is applied to the second axis A2 to perform phase encoding. At the same time, the depth gradient d3 ′ is applied to the third axis A3 to perform phase encoding. Next, at timing t14, the read gradient r1 is applied to the first axis A1 to perform frequency encoding, and the echo E is sampled by the number of read direction data (for example, 256) to collect data for the number of read direction data. . This is the warp gradient w2 ′ and the depth gradient d.
Number of data in warp direction while changing 3 '(for example, 256)
The number of times of data in the depth direction (for example, 32) is repeated to collect data for the number of data for the read direction (for example, 256) × the number of data for the warp direction (for example, 256) × the number of data for the depth direction (for example, 32). Then, from the data, the third axis positive direction images for the number of depth direction data are generated.

Next, at timing t15, the RF pulse R
α is applied and the slice gradient s3 is applied to the third axis A3 to excite protons in the slice region. Next, without applying the flow encode pulse, at timing t17,
A warp gradient w2 'is applied to the second axis A2 to perform phase encoding. At the same time, the depth gradient d3 is applied to the third axis A3 to perform phase encoding. Next, timing t19
Then, while applying the read gradient r1 to the first axis A1 to perform frequency encoding, the echo E is sampled by the number of read direction data (for example, 256), and the data for the number of read direction data is collected. This is repeated for the number of data in the warp direction (for example, 256) × the number of data in the depth direction (for example, 64) while changing the warp gradient w2 ′ and the depth gradient d3, and the number of data for the read direction (for example, 256).
Collect data for the number of data in the warp direction (for example, 256) and the number of data for the depth direction (for example, 64). Then, reference images corresponding to the number of data in the depth direction are generated from the data.

Next, the difference between the first axis forward direction image and the reference image is calculated to generate the first axis flow information image. Also,
The difference between the second axis forward direction image and the reference image is calculated, and the second image
Generate an axial flow information image. Further, the difference between the third axis forward direction image and the reference image is taken to generate a third axis flow information image. Finally, the first axis flow information image, the second axis flow information image, and the third axis flow information image corresponding to the depth direction are combined to generate combined flow information images for the number of depth direction data. This combined flow information image is an isotropic flow information image that is composed of isotropic pixels having the same vertical size and horizontal size and that uses flow information as pixel values. That is, there is no difference in blood vessel extraction ability depending on the flow direction.

The above-mentioned pulse sequence can PS5
, The scan time T required to collect all data is T = (128 × 64 + 128 × 64 + 256 × 32 + 256 × 64) × TR =
40960 · TR. However, 128 x 64 + 128 x 64 + 256 x 32 + 256 x 64
Is the number of warp direction data for obtaining the first axis flow information image × the number of depth direction data and the number of warp direction data for obtaining the second axis flow information image × the number of depth direction data and the third axis flow information image Number of data in warp direction x number of data in depth direction and number of data in warp direction to obtain reference image x number of data in depth direction. The scan time T is shorter than the scan time required to collect all data by the conventional pulse sequence can PS54.

[0057]

According to the anisotropic flow information image processing method and the MRI apparatus of the present invention, the difference in resolution in the vertical and horizontal directions in the anisotropic flow information image is averaged by the image processing operation. It is possible to reduce the difference in blood vessel extraction ability depending on the flow direction. Also, the MR of the flow information of the present invention
According to the imaging method and the MRI apparatus, since the resolution in the radial direction of the blood vessel is always high, it is possible to reduce the difference in the blood vessel extraction ability depending on the flow direction. Furthermore, extension of scan time can be suppressed.

[Brief description of drawings]

FIG. 1 is a block diagram showing an MRI apparatus according to an embodiment of the present invention.

FIG. 2 is an exemplary diagram of a pulse sequence for obtaining two-dimensional PC angiography in the first embodiment.

FIG. 3 is an exemplary diagram of a subject and blood flow.

FIG. 4 is a view showing an example of a first axis flow information image.

FIG. 5 is a view showing an example of a second axis flow information image.

FIG. 6 is a view showing an example of a combined flow information image before image processing calculation in the first embodiment.

FIG. 7 is a view showing an example of a line detection template that emphasizes the vertical direction.

FIG. 8 is a view showing an example of a combined flow information image after image processing calculation in the first embodiment.

9 is an exemplary view of a second axis flow information image after processing the second axis flow information image of FIG. 5 with the template of FIG. 7.

FIG. 10 is another illustration of the composite flow information image in the first embodiment.

FIG. 11 is another exemplary diagram of a line detection template that emphasizes the vertical direction.

FIG. 12 is a view showing an example of a pulse sequence for obtaining two-dimensional PC angiography in the second embodiment.

FIG. 13 is another exemplary diagram of a subject and blood flow.

FIG. 14 is a view showing an example of a first axis flow information image.

FIG. 15 is a view showing still another example of the subject and the blood flow.

FIG. 16 is a view showing an example of a second axis flow information image.

FIG. 17 is another exemplary diagram of the composite flow information image according to the second embodiment.

FIG. 18 is a view showing a modified example of the pulse sequence for obtaining the two-dimensional PC angiography in the second embodiment.

FIG. 19 is an exemplary diagram of a pulse sequence for obtaining three-dimensional PC angiography in the third embodiment.

FIG. 20 is a schematic diagram showing a collection of voxels forming a first axis flow information image in the third embodiment.

FIG. 21 is a schematic diagram showing a collection of voxels forming a second axis flow information image in the third embodiment.

FIG. 22 is a schematic diagram showing a group of voxels forming a third axis flow information image in the third embodiment.

FIG. 23 is a schematic diagram showing a collection of voxels forming a combined flow information image in the third embodiment.

FIG. 24 is a view showing an example of modification of the pulse sequence for obtaining three-dimensional PC angiography in the third embodiment.

FIG. 25 is an exemplary diagram of a pulse sequence for obtaining conventional two-dimensional PC angiography.

FIG. 26 is a view showing an example of a subject and blood flow.

FIG. 27 is a view showing an example of a first axis flow information image.

FIG. 28 is a view showing an example of a second axis flow information image.

FIG. 29 is a view showing an example of a conventional combined flow information image.

FIG. 30 is an exemplary diagram of a pulse sequence for obtaining conventional three-dimensional PC angiography.

FIG. 31 is a schematic diagram showing a collection of voxels forming a conventional flow information image.

[Explanation of symbols]

100 MRI device 6 Display device 7 Computer PS1, PS2, PS3, PS4, PS5, PS54
Pulse sequence g1, g2 line detection template

─────────────────────────────────────────────────── ─── Continuation of front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) A61B 5/055

Claims (6)

(57) [Claims] 1. A method of processing an anisotropic flow information image, comprising anisotropic pixels having different vertical and horizontal sizes and having flow information by angiography as pixel values in an MRI apparatus, said method comprising: According to an angiography in an MRI apparatus, which performs an image processing operation for emphasizing the directionality of the anisotropic pixel having a smaller size relative to the directionality of the anisotropic pixel having a larger size. Anisotropic flow information image processing method. 2. An MRI apparatus configured to generate an anisotropic flow information image, which comprises anisotropic pixels having different vertical and horizontal sizes and has flow information as a pixel value, wherein the size of the anisotropic pixels is Of the anisotropic flow information image, the image processing operation for emphasizing the smaller directivity of the anisotropic flow information image relative to the larger directivity of the anisotropic pixel is provided. Characteristic MRI device. 3. The MRI apparatus according to claim 2, wherein when the anisotropic flow information image has a first axis and a second axis which are orthogonal to each other in real space, a flow encode pulse is added in the first axis direction. An anisotropic first axis flow information image is generated from the collected data, an anisotropic second axis flow information image is generated from the collected data by adding a flow encode pulse in the second axis direction, and the anisotropic first axis flow information image is generated. The first axis flow information image and the anisotropic second
An MRI apparatus characterized in that an anisotropic flow information image is generated by synthesizing with an axial flow information image. 4. When there are a first axis and a second axis which are orthogonal to each other in a real space, a flow encode pulse is added to either the first axis direction or the second axis direction, and the warp direction data number is set from the read direction data number. A data collecting means for collecting data of anisotropic two-dimensional k-space at a minimum, and a first axis flow information image is generated from the collected data by adding a flow encode pulse in the first axis direction and a second axis direction is generated. Flow information image generating means for generating a second axis flow information image from the data collected by adding the flow encode pulse to the first axis flow information image, and the first axis flow information image and the second axis flow information image are combined to generate combined flow information. In the MRI apparatus having a flow information image synthesizing unit for generating an image, the data collecting unit sets a direction to which a flow encode pulse is applied as a warp direction, and is orthogonal to the warp direction. MRI apparatus characterized by collecting data countercurrent as read direction. 5. The MRI apparatus according to claim 4, wherein the composite flow information image is composed of isotropic pixels having the same vertical size and horizontal size in the image. MRI device. 6. When there is a first axis, a second axis and a third axis which are orthogonal to each other in a real space, a flow encode pulse is applied to either the first axis direction, the second axis direction or the third axis direction. Data collecting means for collecting anisotropic three-dimensional k-space data by reducing the number of data in the warp direction from the number of data in the read direction, and the number of data in the depth direction from the data collected by adding a flow encode pulse in the first axis direction. Minute first axis flow information images are generated, and flow encode pulses are added in the second axis direction to generate second axis flow information images for the number of depth direction data from the collected data, and flow encode is performed in the third axis direction. Flow information image generation means for generating third axis flow information images for the number of depth direction data from the data collected by adding pulses, the first axis flow information image, the second axis flow information image, and the above An MRI apparatus having a flow information image synthesizing means for synthesizing a third axis flow information image to generate a synthesized flow information image, wherein the data collecting means applies a flow encode pulse in a first axis direction and Data is collected with the direction being the warp direction, the second axis direction being the lead direction, and the third axis being the depth direction, adding flow encode pulses to the second axis direction, and making the first axis direction the lead direction and the second axis direction the warp. Direction, the third axis direction is the depth direction, data is collected, a flow encode pulse is applied to the third axis direction, the first axis direction is the lead direction, the second axis direction is the warp direction, and the third axis direction depth direction is Increase the number of data in the warp direction to the same as the number of data in the read direction and reduce the number of data in the depth direction according to the increased number of data in the warp direction to collect data. An MRI apparatus characterized by: