JPH09509A - MR image generating method and MRI apparatus - Google Patents

Abstract

(57) [Summary] [Object] To provide an MR image having strong magnetic field inhomogeneity, high S / N ratio, and few motion artifacts. Further, in addition to that, an MR image capable of reducing a ghost due to a phase discontinuity of MR data in k space is provided. [Structure] Two-channel MR data treated as complex number data is collected from an object, the MR data is arranged in a k-space according to an encoding amount, and an MR image is obtained from this arrangement data. At this time, the absolute value image I _abs is reconstructed from the MR data S _l (k _x , k _y ) of the low frequency component lower than the set value on the k space, and the MR of the high frequency component equal to or higher than the set value on the k space is reconstructed. The real component image I _real is reconstructed from the data S _h (k _x , k _y ), and the pixel values of the absolute value image I _abs and the real component image I _real are added for each pixel, and the MR image I (x, y) is added. ) Is generated.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an MR image generating method and an MRI apparatus in magnetic resonance imaging (MRI), and in particular, it mixes absolute value data and real component (real part) data of MR data treated as a complex number. 1
The present invention relates to a hybrid MR image generating method and an MRI apparatus configured to generate an image.

[0002]

2. Description of the Related Art Generally, in magnetic resonance imaging utilizing the magnetic resonance phenomenon of nuclear spins in an object, the collected MR signal is phase-detected by a quadrature detector to obtain a complex number of real data and imaginary data. Processed as data.

An MRI apparatus for collecting MR signals has a problem such as magnetic field inhomogeneity, which is difficult to completely remove in terms of hardware, and causes a phase error of MR signals between pixels.

Since this phase error causes image unevenness, normally, the absolute value of the MR signal which is complex number data is calculated, and an image (absolute value image) depending on the absolute value is displayed to reduce the image unevenness. .

On the other hand, the MR by the spin echo (SE) method
When collecting signals, the effects of magnetic field inhomogeneity described above are largely cancelled. Therefore, even if the image uses only real part data (real component image), if there is no effect of the eddy current, it becomes a spatially sufficiently uniform image, and image unevenness is small. The effects of eddy currents can be reduced by adjusting the hardware or performing phase correction processing by software.

Compared to the absolute value image, the real component image is more likely to be affected by the inhomogeneity of the magnetic field, but has an advantage that the S / N ratio is good and that the artifact (ghost) due to the motion (body motion) is small. is there.

Contrary to the above, the absolute value image is superior to the real component image in magnetic field inhomogeneity, but inferior in S / N ratio (when the original MR signal has a poor S / N ratio). However, there is a problem in that motion artifacts are likely to occur and ghosts due to phase discontinuity in the k-space are likely to occur.

[0008]

As described above, the advantages and disadvantages coexist in both the absolute value image and the real component image, and it is impossible to expect a high quality image and an excellent diagnostic ability with only one of the images.

In particular, signals such as organs and tissues which are important for image diagnosis often come to the central portion (low frequency portion) in the k space, but in the case of a real component image, such low frequency portion is in motion. It has been pointed out that the inconvenience that the influence of artifacts and ghosts is large and that it is likely to cause misdiagnosis. On the other hand, the high frequency components are relatively unaffected by ghost and magnetic field inhomogeneity.

Further, when displaying a real component image, the above-described phase correction processing is actually indispensable, which has contributed to an increase in the calculation load of the signal processing, and in addition to the 0th order and the 1st order. Low-order phase correction was not enough (unevenness cannot be reduced).

The present invention has been made in view of the inconveniences of the above-mentioned conventional MR images, and is strong against magnetic field nonuniformity and has an S / N ratio.
It is an object of the present invention to provide an MR image generating method and an MRI apparatus having a high ratio and less motion artifacts. A further object of the present invention is to provide an MR image generating method and an MRI apparatus capable of reducing a ghost caused by a phase discontinuity of MR data on a k-space.

[0012]

In order to achieve the above-mentioned object, the MR image generating method of the present invention collects 2-channel MR data treated as complex number data from a subject and determines the MR data according to the encoding amount. A MR image having a low frequency component lower than the set value on the k space.
The absolute value image is reconstructed from the data, the real component image is reconstructed from the MR data of the high frequency component above the set value in the k space, and the pixel value of the absolute value image and the real component image is calculated for each pixel. The MR image is generated by addition.

For example, this generation method is an MR image generation method applied to MR imaging using a scan method in which different MR echo data is arranged in each of a plurality of segment areas obtained by dividing the k space, The low frequency component and the high frequency component are separated corresponding to the division position of the segment area. For example, the scanning method is a high speed SE method.

Further, in a preferred mode, the processing for reconstructing the absolute value image cuts out the MR data of the low frequency component from the k space, and zero-fills the position on the k space for the non-cut out component. Setting is performed to form matrix-size first k-space data, and Fourier transform of the first k-space data is performed. The process of reconstructing the real component image is performed by the MR of the high-frequency component.
Data is cut out from the k-space, and zero-filling processing is performed on positions on the k-space for components not cut out to form second k-space data of a set matrix size,
This is a process of Fourier-transforming the second k-space data.

In order to achieve the above object, the MRI of the present invention
The apparatus has means for collecting 2-channel MR data treated as complex number data from a subject, and means for arranging this MR data in k space according to the encoding amount. And a means for reconstructing an absolute value image from MR data of a low frequency component lower than a set value on the k space, and MR data of a high frequency component above the set value on the k space. And a means for reconstructing the real component image from the above and a means for generating the MR image by adding the pixel values of the absolute value image and the real component image for each pixel.

[0016]

According to the MR image generating method and the MRI apparatus of the present invention, 2-channel MR data treated as complex number data is collected from the subject, and this MR data is arranged in the k space according to the encoding amount. An MR image is created from this arrangement data. Specifically, the absolute value image is reconstructed from the MR data of the low frequency component lower than the set value on the k space, and the real component image is reconstructed from the MR data of the high frequency component above the set value on the k space. Composed,
The MR image is generated by adding the pixel values of the absolute value image and the real component image for each pixel.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

A schematic configuration of a magnetic resonance imaging (MRI) apparatus according to this embodiment is shown in FIG. This magnetic resonance imaging apparatus includes a magnet unit for generating a static magnetic field, a gradient magnetic field unit for adding position information to the static magnetic field, a transmitter / receiver unit for selective excitation and MR signal reception, system control and image reconstruction. It is equipped with a control / arithmetic unit.

The magnet section includes, for example, a superconducting magnet 1,
A static magnetic field power supply 2 for supplying a current to the magnet 1 is provided, and a static magnetic field H ₀ is generated in the Z-axis direction of the cylindrical diagnostic space into which the subject P is inserted.

The gradient magnetic field section is composed of the X,
Three sets of gradient magnetic field coils 3x to 3z in the Y and Z-axis directions, a gradient magnetic field power supply 4 that supplies a current to the gradient magnetic field coils 3x to 3z, and a gradient magnetic field sequencer 5 that controls the power supply 4.
With. This sequencer 5 has a computer,
A signal for instructing the acquisition sequence (see FIG. 3) of the gapless type multi-slice imaging related to the high-speed SE method, for example, is received from the controller 6 (installed with a computer) of the entire apparatus. As a result, the gradient magnetic field sequencer 5 controls the application and strength of each gradient magnetic field in the X, Y, and Z axis directions according to the commanded sequence, and these gradient magnetic fields can be superimposed on the static magnetic field H _0. There is. In this embodiment, the gradient magnetic field in the Z-axis direction out of the three axes orthogonal to each other is the slice gradient magnetic field G _S , the gradient magnetic field in the X-axis direction is the reading gradient magnetic field G _R, and the gradient magnetic field in the Y-axis direction is the phase. The encoding gradient magnetic field G _E is used.

The transmission / reception unit includes a high-frequency coil 7 arranged near the subject P in the imaging space inside the magnet 1, a transmitter 8T and a receiver 8R connected to the coil 7, and a transmitter 8T. And RF for controlling the operation timing of the receiver 8R
A sequencer 9 (with a computer). Under the control of the RF sequencer 9, the transmitter 8T and the receiver 8R supply the RF current pulse of the Larmor frequency for exciting the nuclear magnetic resonance (NMR) to the high frequency coil 7, while the high frequency coil 7 receives the RF current pulse. The MR signal (high frequency signal) is subjected to various kinds of signal processing to form image data of a digital signal, the details of which are shown in FIG.

In addition to the controller 6 described above, the control / arithmetic unit inputs the image data formed by the receiver 8R and reconstructs the image data.
A storage unit 11 for storing the reconstructed image data, a display 12 for displaying an image, and an input device 13 operated by an operator. Arithmetic unit 10
Specifically, processing such as arrangement of actually measured data in a two-dimensional Fourier space which is a memory space and Fourier transform for image reconstruction is performed. The controller 6 controls the operation contents and operation timing of the gradient magnetic field sequencer 5 and the RF sequencer 9 while synchronizing them.

The transmitter 8T and the receiver 8R are specifically formed as shown in FIG. Among them, the transmitter 8T includes oscillators 20 and 2 having oscillation frequencies f0 and Δf.
1 and includes a phase selection unit 22, a frequency conversion unit 23, an amplitude modulation unit 24, a high frequency power amplification unit 25, and a modulated wave generation unit 26, which are sequentially arranged on the output side of the oscillation unit 20. .

The modulated wave generator 26 is adapted to generate a modulated wave F (t) of, for example, a SINC function corresponding to a predetermined slice thickness when a control signal arrives from the RF sequencer 9. The modulated wave F (t) is supplied to the amplitude modulator 24.

The phase selector 22 which has received the oscillation signal of the oscillator 20 whose frequency is f0 selects the phase of the signal as φ and sends it to the frequency converter 23 in the subsequent stage. Frequency converter 23
An oscillation signal of frequency Δf is input to the other oscillator 21. Therefore, the frequency conversion unit 23 forms a high frequency signal having a frequency f0 ± Δf using the two input signals, and supplies one of the high frequency signals having a frequency f0 + Δf to the amplitude modulation unit 24.

The amplitude modulation section 24 modulates a signal (phase φ) having a frequency f0 + Δf, which is a carrier, with a modulation wave F (t), and passes it through a high frequency power amplifier 25 for power amplification.
The high frequency coil 7 in the magnet 1 is supplied.

On the other hand, the receiver 8R has a preamplifier 40 connected to the high frequency coil 7, an intermediate frequency converter 41 and a phase detector 4 sequentially connected to the output side of the preamplifier 40.
2, low-frequency amplifier 43, low-pass filter 44, and A
It has a -D converter 45. The high frequency signal of NMR amplified by the preamplifier 40 is converted into an intermediate frequency by the intermediate frequency converter 41 and supplied to the phase detector 42. The phase detection unit 42 detects two MR signals, which can be handled as a complex number signal and are out of phase by 90 degrees, from the input signal. The two detected signals are sent to the 2-channel A / D conversion section 45 via the 2-channel low frequency amplification section 43 and the low-pass filter 44. A-D converter 4
The MR signal converted into a digital signal in 5 is read into the arithmetic unit 11.

Next, the function and effect of this embodiment will be described.

First, when the gradient magnetic field sequencer 5 and the RF sequencer 9 are instructed by the controller 6 to acquire an acquisition sequence of the high-speed SE method for gapless multi-slice imaging, as shown in FIG. The application of the gradient magnetic field to the sample P and the transmission / reception of the high frequency signal are controlled. Note that the number of echoes of the fast SE method in the "5", the k-space as shown in FIG. 4, the phase-encoding k _y segment area 5 divided by equally dividing point direction S
It is assumed that each encode signal is arranged in each of G1 to SG5.

First, the slicing gradient magnetic field G _S is applied from the gradient magnetic field power source 4 through the gradient magnetic field coils 3 _z , 3 _z , and when the gradient magnetic field G _S rises to a constant value, the transmitter 8T and A 90 ° RF pulse (selective excitation pulse) is applied only once via the high frequency coil 7. As a result, the imaging region having a predetermined slice width forming one of the multi-slice planes of the subject is selectively excited, and the nuclear spins in the plane are flipped to the y ′ axis (rotational coordinate).

Then, a phase-encoding gradient magnetic field G _E corresponding to each encoding position of SG1, for example, the first segment area in a predetermined number of segment areas (here, 5 divisions are adopted) on the k-space is the gradient magnetic field power supply. 4 is applied to the subject P through the gradient magnetic field coils 3y and 3y. As a result, the nuclear spins in the slice plane rotate by the phase corresponding to the position in the encoding direction.

Then, similarly to the above, the first 180 ° RF pulse P ₁ (refocus pulse) is applied together with the slicing gradient magnetic field G _S. As a result, the nuclear spin rotates 180 degrees around the y ′ axis, and then the first spin echo signal S ₁ is generated. At this time, the read gradient magnetic field G _R is applied via the gradient magnetic field coils 3x, 3x, and the first spin echo signal S ₁ passes through the high frequency coil 7 during the rising period of the gradient magnetic field G _R. To be collected.

After that, a phase encoding gradient magnetic field G _E corresponding to the encoding position in the second encoding segment region, for example, SG2 is applied. Then, similarly to the above, the second 180 ° RF pulse P ₂ is applied together with the slice gradient magnetic field pulse G _S , and the second spin echo signal S ₂ is generated. This second spin echo signal S ₂ is collected via the high frequency coil 7 together with the application of the read gradient magnetic field G _R. Further, similarly, the third to
The fifth spin echo signals S _{3 to} S ₅ are also segment regions S
Collected for G3 to SG5, respectively.

The above sequence for one 90 ° RF pulse is repeated for one shot.

The spin echo signals S ₁ , S ₂ , ... S ₅ received by this series of sequences are sequentially received by the receiver 8R.
Is subjected to amplification, intermediate frequency conversion, phase detection, and low frequency amplification, and is then converted into a complex digital signal and output to the arithmetic unit 10.
In the arithmetic unit 10, the echo signal is arranged in each of the segment areas SG1 to SG5 of the k space according to the phase encode amount.

The arithmetic unit 10 has a C installed in advance.
The software function of the PU performs the processing partially shown by the one-dot chain line shown in FIG.

That is, of the two-dimensional echo data arranged according to the phase encoding amount in the k space, the echo data S _l (k _x , k) of the central segment area SG3 including the zero encoding position k _y = 0. _y ) is cut out,
Remaining segment areas SG1, SG2, SG4, SG5
Echo data S _h (k _x , k _y ) on the k space corresponding to
Zeros (= 0) are filled in the positions of (2) to newly create two-dimensional data in the k space (step S1 in FIG. 4). At this time, an appropriate window may be set in order to remove the truncation artifact.

Next, the arithmetic unit 10 carries out step S1.
The two-dimensional echo data on the k-space newly created in 2 is subjected to two-dimensional Fourier transform, and the real-time image V _l (x, y) is obtained.
Reconfigure. This image V _l (x, y) is real part data,
It consists of two channels of imaginary part data.

Next, the arithmetic unit 10 calculates the absolute value of the data of each pixel of the reconstructed two-dimensional image V _l (x, y), and the absolute value image I _abs = | V _l (x, y). | Is created (step S3).

On the other hand, the arithmetic unit 10 has four segment regions SG1, SG2, SG4, S other than the central portion.
For each of the echo data S _h (k _x , k _y ) of G5, only the real component is obtained, so that steps S4 to
The process of S7 is sequentially performed.

Firstly, the echo data _{S h (k x, k y} ) for each of the zero-order, first-order low-order phase correction is performed (step S4). This compensates for a relatively large phase error between pixels. Note that this phase correction does not necessarily have to be performed. Further, when the increase of the calculation load does not cause any particular problem, the DC component removal processing can be performed at this stage in combination with the phase correction or alone.

Further, the segment areas SG1, SG2,
The echo data S _h (k _x , k _y ) of SG4 and SG5 are cut out, and new zero-filled two-dimensional echo data on the k space is created at the position of the middle segment area SG3 by calculation (step 5). The new echo data is then two-dimensional Fourier transformed and reconstructed into the other two-dimensional image V _h (x, y) in real time (step S7). This image V _h (x, y) is also composed of two channels of real part data and imaginary part data due to the above-mentioned quadrature detection.

Therefore, the image V _h composed of this complex number data
Only the real part data is selected from the (x, y) data (step S7). As a result, the real component image I _real
= “Real part of V _h (x, y)” is obtained.

When the absolute value image I _abs real component image I _real is created in this way, the arithmetic unit 10 finally adds the pixel values of both images I _abs and I _real for each pixel to obtain the final two-dimensional image. MR image I (x, y) = | V _l (x, y) | +
The "real part of V _h (x, y)" is obtained.

[0045] As mentioned above, to reconstruct the absolute value image for the heart of the phase encoding k _y direction (low-frequency portion) in the k-space of arranging the echo data in the present embodiment, the peripheral part (high frequency part) For, the real component image is reconstructed, and both are added pixel by pixel to form the final image. As a result, the central portion on the k-space, which is dominant in determining the global density distribution of the image, becomes the absolute value image, and the advantage of the absolute value image can be inherited. On the other hand, the peripheral part on the k space becomes a real component image, and the advantages of the real component image itself can be inherited. Therefore, the MR hybridized in this way
According to the image, the magnetic field inhomogeneity is stronger than that of the conventional image having only the real component, so that the phase error is small, the density unevenness is improved, and the S / S is smaller than that of the conventional image having only the absolute value. Since the N ratio is good and the motion artifacts are small, the image distortion is also reduced, which is a contradictory advantage in the related art.

Further, in the case of the k-space division scan using the high-speed SE method as the pulse sequence for imaging as in the above embodiment, the ghost due to the phase error due to the discontinuity at the echo connection in the k-space Can be reduced,
This point also contributes to image distortion improvement. The advantage of reducing the ghost during the divided scan can be obtained similarly when the GRASE method, the EPI method, or the like is used as the imaging pulse sequence.

Further, in the case of a real component image as in the prior art, phase correction was indispensable for all pixels, but according to the present embodiment, it is applied to echo data in the peripheral part on the k space. Since it can be performed only when necessary, the calculation load after echo data collection can be significantly reduced.

Further, in the present embodiment, each of the two-dimensional echo data created before the processing of the Fourier transform is padded with zeros at the encoding position where no data is contained so that the matrix size in the k space becomes the same. . For this reason, the matrix sizes of the reconstructed images after the Fourier transform are the same as each other, and the image interpolation process is not necessary. Of course, it is also possible to perform the Fourier transform with the clipped size without performing the zero padding, and set the same matrix by interpolation processing in the reconstructed image thereafter.

The MR image generation method and MR of the present invention
The I apparatus is not limited to the case of using the above-described high-speed SE method, but can be applied to a method and apparatus for arranging echo data in k space and reconstructing an MR image from this arrangement data.
In the k-space division scan by the high-speed SE method as described above, it is preferable to separate the low-frequency region and the high-frequency region in the k-space by using the break of the segment area according to the echo signal. If not employing the k space division scanning method, such low frequency range and high frequency range may be separated at any position for each of the phase encoding k _y direction and read k _x direction. Usually, the data continuity in the k-space toward the phase encoding k _y direction is lower than the read k _x direction (frequency encode direction), it is preferable to divide the phase encoding direction.

Further, the number of echoes when the high speed SE method is used is not limited to the above-mentioned "5 echoes", and is arbitrary.

Furthermore, when the fast SE method is used, the order of collecting the generated spin echo signals and the order of the segment areas in the k space in which the spin echo signals are arranged are not necessarily those of the above-described embodiment. However, the first spin echo signal S1 may be arranged in the central segment region SG3 in accordance with the encoding amount.

[0052]

As described above, according to the present invention,
Of echo data on the k space, the low frequency component of the central part is reconstructed as an absolute value image, the high frequency component of its peripheral part is reconstructed as a real component image, and both are added for each pixel to obtain the final MR. By adopting a hybrid configuration method for images, it is possible to obtain the effects of strong magnetic field inhomogeneity, reduction of image unevenness, good S / N ratio, less motion artifacts, and reduction of image distortion. .

[Brief description of drawings]

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

FIG. 2 is a block diagram of a transmitter and a receiver according to the same embodiment.

FIG. 3 is a pulse sequence diagram showing an example of a fast SE method for imaging.

FIG. 4 is a flowchart-like explanatory diagram showing a state of a segment area in a k-space and processing by a hybrid configuration in an arithmetic unit.

[Explanation of symbols]

 1 magnet 2 static magnetic field power supply 3x, 3y, 3z gradient magnetic field coil 4 gradient magnetic field power supply 5 gradient magnetic field sequencer 6 controller 7 high frequency coil 8T transmitter 8R receiver 10 arithmetic unit

Claims (5)

[Claims] 1. An MR image generation method for collecting 2-channel MR data treated as complex number data from a subject, arranging the MR data in a k-space according to an encoding amount, and obtaining an MR image from the arrangement data. In, the absolute value image is reconstructed from the MR data of the low frequency component lower than the set value on the k space, and the real component image is reconstructed from the MR data of the high frequency component above the set value on the k space. A method for generating an MR image, comprising adding the pixel values of the absolute value image and the real component image for each pixel to generate the MR image. 2. An MR image generation method applied to MR imaging using a scan method in which different MR echo data are arranged for each of a plurality of segment areas obtained by dividing the k space, wherein the low frequency component and the high frequency component are used. The MR image generating method according to claim 1, wherein the components are separated corresponding to division positions of the segment areas. 3. The MR image generating method according to claim 2, wherein the scanning method is a high speed SE method. 4. The process of reconstructing the absolute value image cuts out the MR data of the low-frequency component from the k-space, and zero-fills the position on the k-space for the uncut-out component to set a matrix. The first k-space data having a size is formed, and the first k-space data is Fourier-transformed. In the process of reconstructing the real component image, the high-frequency component MR data is cut out from the k-space. At the same time, it is a process of zero-filling a position on the k-space with respect to a component that is not cut out to form second k-space data of a set matrix size, and Fourier-transforming the second k-space data. The described MR image generation method. 5. A means for collecting 2-channel MR data treated as complex number data from a subject, and a means for arranging this MR data in a k-space according to the encoding amount, and MR is calculated from this arrangement data. In an MRI apparatus adapted to obtain an image, means for reconstructing an absolute value image from MR data of a low frequency component lower than a set value on the k space, and high frequency components of the set value or more on the k space. An MRI apparatus comprising: means for reconstructing a real component image from MR data; and means for generating the MR image by adding pixel values of the absolute value image and the real component image for each pixel. .