JPH0321224A - Magnetic resonance imaging - Google Patents

Abstract

PURPOSE:To eliminate aliasing artifact and restore image of high pictorial quality at a high speed by performing a saturation sequence of applying phase encode magnetic field for saturation and a pair of RF pulses for saturation of different sending frequencies before signal collection sequence, and exciting area outside the field of view in the phase encoding direction. CONSTITUTION:A computer 16 calculates intensities of slicing field GS, signal reading field GR, and phase encoding field GP based on respective input values and issues sending frequencies of RF pulses A1 and A2 corresponding to respective slices. It also calculates the intensity of encoding field GP'' for saturation, obtains sending frequencies fs1 and fs2 of RF pulses A11 and A12 for saturation with respect to phase encoding direction, and sets a first phase encoding amount for signal collection sequence. A sequence control device 15 performs the signal collection sequence according to magnetic field control data for both saturation and signal collection sequences transmitted from the computer 16, and collect all the specified image data.

Description

[Detailed Description of the Invention] [Industrial Application Field] This invention relates to magnetic resonance imaging that uses NMR (nuclear magnetic resonance), ESR (electron spin co-III), etc. to construct spectra and tomographic images of a subject. This article relates to a magnetic resonance imaging method that can eliminate aliasing artifacts (aliased artifacts) that occur when the subject is larger than the imaging field of view, regardless of differences in signal acquisition sequences, without deteriorating spatial resolution. ．． [Prior Art] Conventionally, a high-frequency magnetic field pulse (hereinafter referred to as RF pulse) perpendicular to the static magnetic field and a gradient magnetic field along three orthogonal axes (X Y Z) parallel to the static magnetic field have been applied to a subject in a static magnetic field. A magnetic resonance imaging apparatus is well known that acquires a predetermined number of image data based on magnetic resonance signals induced from the subject by applying .

FIG. 4 is a block diagram showing a general magnetic resonance apparatus. In the figure, a subject (1) such as a human body is placed on an examination table in a static magnetic field generator (2) that generates a static magnetic field in the Z-axis direction. (3
〉 RF for the object (1)
Apply pulse A (high frequency energy) and test object (1)
A high-frequency coil (4) receiving a magnetic resonance signal (for example, an NMR signal) B from the transmitter (7) and a receiver (8) is connected via a matching device (5) and a transmitting/receiving switching device (6). A tilted magnetic basin G R in three axes orthogonal to the test object (1)
．． Gp and G.P. Gradient magnetic field coil <9>, (1
1) and (13) are each gradient magnetic field power supply (10>, (l2
) and (14). Here, the X-axis gradient magnetic field is used as the signal readout magnetic field for frequency encoding OR
, the Y-axis gradient magnetic field is a phase encoding magnetic potion G p that provides a phase encode amount, and the Z-axis gradient magnetic field is a slice magnetic potion G1 for specifying a tomographic plane. The sequence control device (15) includes an examination table (3), a transmitter (7), a receiver (8), each gradient magnetic field power source (10), (1
The entire device including 2) and (14) is controlled in a predetermined sequence. Receiver (8) and sequence control device (15)
The computer (16) connected to Process resonance signal B. An image display device (18) for displaying tomographic images is connected to the operation console (17).Next, while referring to the pulse sequence diagram in FIG. The resonance imaging method will be explained.The pulse sequence shown in FIG. 5 is stored in advance as part of a program in the computer (16) in FIG. 4, and is executed by the sequence control device (15). In addition, here, magnetic resonance signal B is obtained using the Subin echo method.
A case is shown in which a tomographic image is constructed using the two-dimensional Fourier transform method from the image data based on the magnetic resonance signal B. First, insert the test tube (1) into the high frequency coil (4) and each gradient magnetic field coil (9>, (11) and <13>), and
and a high-frequency magnetic field pulse with a selective frequency (
RF pulse)^1 and slice magnetic field 0.1 to the subject (1
3). This RF pulse ^1 usually has a flip angle of 9G" and is called a 90° pulse. As a result, high-frequency energy is supplied to the nuclear spins in the desired cross-sectional plane in the object (1), and the RF pulse ^1 The phase of the spins begins to be disturbed from the center (peak) position of The pulse area (shaded area) is applied so that it matches the pulse area of 0.1 slice magnetic field after the peak of RF pulse A1.Next, the Y-axis gradient magnetic field coil (11) is driven to generate a phase encode magnetic field. Gp is applied to disturb the phase of the spin in the Y-axis direction within the tomographic plane, and
Drive the axial gradient magnetic field coil (9) to generate a signal readout magnetic field G
, 11 are applied. Next, after applying an RF pulse ^2 with a flip angle of 1800, a signal readout magnetic field G*. of the same polarity as Gal is applied. 1
While applying the magnetic resonance signal (Subin echo signal) B
Receive. This magnetic resonance signal B is transmitted to the high frequency coil (4
) from the transmitter/receiver switching device (6) and the receiver (8>) into the computer (16). In this signal collection sequence, the pulse areas shown by the shaded areas of each signal readout magnetic field Gel and Gs2 coincide. In other words, when the echo time TE has elapsed from the peak time of the 90° pulse A1, the magnetic resonance signal B reaches its peak value.The timing at which the magnetic resonance signal B reaches its peak is determined by the signal readout magnetic field Gll and the RF Depending on the application timing of pulse A, the 180° pulse ^2 is applied so that it reaches its peak at the time when 172 of the echo time TE has elapsed from the peak time of the 90° pulse ^1.Then, the image data is The magnetic resonance signal B is collected by sampling the magnetic resonance signal B at a predetermined number of sampling points during the application of the signal readout magnetic field GR2.
While sequentially changing the phase encode amount corresponding to the pulse area at a predetermined pitch (see broken lines and arrows), the process is repeated a number of times (signal collection number) corresponding to a predetermined number of pixels N (for example, 256). Therefore, when the number of pixels of the final elliptical tomographic image is N be.

After executing the signal acquisition sequence N times, the computer (16) performs a two-dimensional Fourier transform on the pulse train of image data based on the magnetic resonance signal B to reconstruct a tomographic image with a desired matrix size NXN, and image Display device (18)
Display on . For example, the strength of the phase encoding magnetic field Gp that changes for each sequence is expressed as follows: where the nuclear magnetic resonance ratio is γ, the imaging field of view in the phase encoding direction (hereinafter simply referred to as the field of view) is LP, and the matrix size in the phase encoding direction is Np. γLpSGp-dt=2-π...■However,
Import = Np/2, NP/2-1,..., 1,O,-
1,..., -Np/2+1 is sequentially changed. Therefore, the amount of change ΔGp in the phase encode magnetic field Gp
is SΔG1dt=2π/(γ・2LP)...■.

The computer <16) performs two-dimensional Fourier transform on the imaging data based on the plurality of magnetic resonance signals B, and the matrix size is N.
A tomographic image of a desired slice plane of +txNp is then extracted again. Usually, the matrix size NR in the frequency encoding direction is equal to the matrix size N in the phase encoding direction.

However, in general, the size of the subject (1) is determined by the field of view N. If the size of the subject in the phase encoding direction (Y-axis direction) is larger than the field of view L as shown in Fig. 6, aliasing as shown in Fig. 7 may occur due to the folded image of the non-imaging part. Artifact B' (shaded area) occurs. This is because when the l80° pulse is applied, spins outside the slice region in the phase encoding direction are also excited and contribute to the magnetic resonance signal B. In order to remove such aliasing artifacts, a method can be considered in which the matrix size Np and the field of view Lp are set large, and the image is captured, and the imaged data for the matrix size and field of view at the center is acquired. However, in this case, as is clear from equation (2), the minimum amount of change ΔGp of the phase encode magnetic field G9 becomes small, so the influence of the noise signal increases and the spatial resolution deteriorates. Also, as shown by the broken line in Figure 5, the 180° pulse^2
It is also conceivable to apply a phase encode magnetic field GP' at the time of application of , and further slice in the phase encode direction, thereby preventing excitation in areas other than the sliced region.

However, this method cannot be applied to a multi-slice method in which the signal acquisition sequence is repeated while changing the frequency of the RF pulse A because it affects other slice planes. Further, it cannot be applied to the gradient field echo method, which achieves high speed by applying the RF pulse A with a low flip angle, because the 180'' pulse is not used.

[Problems to be Solved by the Invention] As described above, in the conventional magnetic resonance imaging method, if the subject (1) exists outside the visual field L in the phase encoding direction, it is difficult to distinguish it from the phase encode amount within the visual field LP. Therefore, there is a problem in that the folded image overlaps the image within the field of view L, resulting in so-called aliasing artifacts.

In addition, to prevent this, the field of view Lp and matrix size N
When p is increased, the minimum amount of change ΔGe becomes smaller, resulting in a problem that the spatial resolution deteriorates.

Furthermore, the method of applying the phase encode magnetic field G p' (see broken line in Figure 5) when applying the 180'' pulse^2 has the problem that it cannot be applied to the multi-slice method, gradient field echo method, etc. This invention was made to solve the above-mentioned problems, and it eliminates aliasing artifacts without degrading spatial resolution, regardless of differences in signal acquisition sequences, and allows high-quality images to be produced at high speed. The purpose of this study is to obtain magnetic resonance imaging that can be reconstructed.

[Means for Solving the Problems] The magnetic resonance imaging method according to the present invention is characterized in that, before executing a signal acquisition sequence, a saturation phase encoding magnetic field and a pair of saturation RF pulses having different transmission frequencies are applied. The sequence is executed to excite the area outside the field of view in the phase encoding direction that is the imaging target. [Operation] In the present invention, spins outside the field of view are selectively saturated in advance before a signal acquisition sequence is executed, and for any signal acquisition sequence such as a multi-slice method or a gradient field echo method, Prevent aliasing artifacts from occurring without degrading spatial resolution. As a result, when the field of view in the phase encoding direction is small, the number of phase encodes can be reduced and speeding up or an improvement in the S/N ratio can be realized. [Embodiment] Hereinafter, an embodiment of the present invention will be described with reference to the drawings, taking as an example the case where the magnetic resonance signal B is received by the spin echo method and the image is reconstructed by the two-dimensional Fourier transform method, as described above. The magnetic resonance apparatus to which this invention is applied is shown in FIG. 4, and the signal acquisition sequence is the conventional one. Therefore, it is only necessary to replace some of the programs and memory in the sequence control device (15) or the computer (16) shown in Fig. 4, or to add a new program. Fig. 1 shows an embodiment of the present invention. Pulse sequence diagram, Fig. 2 shows the sequence control device (15) in this invention.
) and the operation of the computer (16). Moreover, FIG. 3 is an explanatory diagram showing the spin state of the imaging field of view Lp and the saturated region LP' in the subject (1), and as shown in FIG. The case where N*xNp) is shown. Note that this example shows the case where the multi-slice method is applied as the signal acquisition sequence. First, the number of multi-slices n,
Slice thickness d, slice pitch ΔS, phase encoding direction field of view L. and frequency encoded field of view L7,
Perform initial settings (step S1). The calculator (16) is
Based on each input value, the slice magnetic field G. , the intensities of the signal readout magnetic field G8 and the phase encode magnetic field G are calculated (step S2), and the transmission frequencies of the RF pulses ^1 and ^2 corresponding to each slice of the multi-slice are derived (step S3). Furthermore, the intensity of the saturation phase encoding magnetic field Gp' is calculated (step S4), and the saturation RF pulses ^11 and ^ in the phase encoding direction are calculated.
12 transmission frequencies fst and fsl are determined (step S
5). Furthermore, an initial phase encode amount for the signal acquisition sequence is set (step S6). Note that the saturation RF pulse ^11 derived in step S5
The transmission frequencies fs'1 and rs2 correspond to the amount of frequency shift from the transmission frequency fo of the RF pulses ^1 and A2 during the signal collection sequence, fst=(?'/2ff)・Gp"(L=/2 )+fs
inc...-■fs4==fsl
・ ...Represented by ■. However, γ is the nuclear magnetic resonance ratio, G1 is the phase encoding magnetic field for saturation, Lt is the field of view in the phase encoding direction, and fsinc is the RF pulse for saturation ^11 and ^12? The siM frequency, Lp'' is the size of the saturation region in the phase encoding direction, and the saturation phase encoding magnetic field Gp'' is expressed as Gp''=4r・fsinc/rLp.Hereinafter, the sequence control device (15) is , calculator (16)
Based on the magnetic field control data for the saturation sequence and the signal acquisition sequence transmitted from the controller, the signal acquisition sequence shown in Fig. 1 is executed, and all predetermined image data are acquired. In addition, in the initial state before applying the RF pulse and gradient magnetic field, the spins within the object (1) are as shown by the arrow in FIG. 3(a), regardless of the imaging field of view L. and the saturation region Lp'' They are in a ripe equilibrium state following the static magnetic field, and are all oriented in the vertical direction.Here, the saturated regions L'' and LP2'' are excited (step S7), and as shown in Figure 3(b), each saturated region is In other words, while applying the saturation phase encode magnetic field GP'', the saturation RF pulse ^11 (90'' pulse) having the transmission frequency fsl of the formula (■) is applied. , first saturation region L
．． The spin of the second saturation region L, " is tilted by 90 degrees, and then the saturation RF pulse A12 (90 degrees pulse) of the transmission frequency fs2 of formula (2) is applied to tilt the spin of the second saturation region L, " by 90 degrees. Thereafter, by further applying the saturation phase encoding magnetic field GP'' (shaded area), the phases of the spins in the saturation regions L.'' and LP2'' are disturbed ( Dephasing (called dephase) makes the longitudinal division of the spins very small as a saturated state. After the saturation sequence is thus completed, a normal signal acquisition sequence is performed to collect the magnetic resonance signal B of each slice. (Step S8) At this time, the spins within the imaging field of view LP and the saturation region L,'' are changed to 90 by the RF pulse ^1.
``However, only the spins in the imaging field of view Lp contribute to the later received magnetic resonance signal B (imaging), and the saturated region L
The spin of P'' is the magnetic resonance signal B as shown in Fig. 3(e).
It makes little contribution to Therefore, in subsequent imaging,
Images in <saturated regions t, p-> other than the phase encoding visual field L do not overlap within the visual field LP. In this case, since it is a multi-slice method, the RF pulse ^
The magnetic resonance signal B is repeatedly received while the transmission frequencies of 1 and ^2 are shifted in accordance with the slice pitch ΔS.

Also, along with the 180° pulse ^2, the slice magnetic field Gs1.2
is applied to avoid exciting spins other than the target slice, and in order to equalize the phase disturbance of the spins due to the shaded area of the 90° pulse ^1, the 180° pulse ^2 is made longer by the same pulse area (the shaded area). is being applied.

Usually, the longitudinal relaxation time T1 of the spin to be imaged is 1 second to several 1
001 seconds, which is longer than each repetition time (several tens of milliseconds) in a multi-slice signal acquisition sequence, so as shown in the figure, multiple signal acquisitions are possible after one saturation sequence. Therefore, the saturation sequence may be executed again before several 100+ seconds have passed after the execution of the saturation sequence. Then, it is determined whether or not the signal collection for the phase encode matrix N, has been completed (step S9). If not, the next phase encoder code amount is set (step StO), and then step S7
Return to If it has been completed, the image of each slice is re-elliptized (step S11) and displayed on the image display device (18) (step 7'S12), and the imaging pulse sequence is ended.

As a result, by simply changing the program in the computer (16),
A desired tomographic image with suppressed aliasing artifacts is constructed based on the signal acquisition sequence as needed. At this time, the imaging time T required for the signal acquisition sequence required for imaging is expressed as T=TRxNp, where TR is the repetition time. Therefore, depending on the size of the tomographic image area required for imaging, the size of the phase encoding field of view LP is adjusted to the frequency encoding direction field of view Ll. Set smaller than l,
Moreover, the imaging time T can be shortened by reducing the phase encoding matrix N by the amount that the phase encoding field of view Lp is made smaller. For example, compared to the usual case where each matrix size Np and NII are set as N-=N*, it is possible to reduce the matrix size by a factor of (L,/L,l). At this time, from equation (2), the relationship between the maximum value G, wax of the phase encoding magnetic field G2, the phase encoding visual field Lp, and the matrix size NP is as follows: 7 L p 5 G pmax-dt= N px
−■, the maximum phase encoding magnetic field G pl
naX is set to satisfy the formula (■). Further, in the shortened imaging time T, if the signal acquisition sequence is further repeated and image data is added, the S/N ratio of the tomographic image can be improved. In the above embodiment, the phase encoding magnetic field G1 for saturation was applied for a long time by the shaded area to disturb the phase of the saturation region L1, but in the normal signal acquisition sequence that is subsequently executed, the phase encoding magnetic field G9 is applied. Since the phase of the spins in the saturation region L1 is disturbed, the same effect can be achieved even if the shaded portion of the saturation phase encode magnetic field G1 is not applied.

In addition, although we have shown the case where the magnetic resonance signal B is a Subin echo signal, it is a gradient field echo signal (gradient magnetic field echo signal) received without applying a 180''' pulse^2 and a slice magnetic field of 0.12, or an electron spin It may also be a resonance signal, etc.For example, even in high-speed imaging methods such as the gradient field echo method using RF pulses with a small tilt angle (low flip angle), the field of view Lp is By saturating the outside, an image without aliasing artifacts can be obtained in the same way as described above. Also, the transmission frequency f of the saturation RF pulses ^l1 and ^12
s. and fst, the center frequency f in the phase encoding direction
o, but each transmission frequency fsl and fs. By setting arbitrarily, it is possible to obtain images without aliasing artifacts not only in the central part of phase encoding but also when tomographic images shifted from the center are to be imaged. Further, although the case where the image is constructed using the two-dimensional Fourier transform method has been described, the same effect can be achieved even in the case of the three-dimensional Fourier transform method. Further, although the case where a tomographic image is constructed based on the magnetic resonance signal B has been shown, the present invention can also be applied to the case of an imaging method in which a specific magnetic resonance signal B is collected to obtain a spectral orescope. [Effects of the Invention] As described above, according to the present invention, before executing a signal acquisition sequence, a saturation sequence is executed in which a saturation phase encode magnetic field and a pair of saturation RF pulses having different transmission frequencies are applied. Since the area outside the field of view in the phase encoding direction to be imaged is excited and saturated in advance, the spatial resolution can be improved for any signal acquisition sequence such as the multi-slice method or the gradient field echo method. It is possible to prevent the occurrence of aliasing artifacts without deteriorating the image quality. Furthermore, when the field of view in the phase encoding direction is small, the number of required phase encodes can be reduced, making it possible to increase speed or improve the S/N ratio. Therefore, using a conventional system including a sequence control device and a computer, it is possible to obtain a magnetic resonance imaging method that can quickly restore high-quality images with suppressed aliasing artifacts without increasing costs. be.

[Brief explanation of the drawing]

FIG. 1 is a pulse sequence diagram showing one embodiment of the present invention, FIG. 2 is a flowchart diagram for explaining the operation of imaging by the pulse sequence of FIG.
An explanatory diagram showing the spin state in the saturation region that changes depending on the saturation sequence in the figure, FIG. 4 is a block diagram showing a general magnetic resonance apparatus, and FIG. 5 is a pulse sequence diagram showing the conventional magnetic resonance imaging method. FIG. 6 is an explanatory diagram showing the field of view in the phase encoding direction and the frequency encoding direction, and FIG. 7 is an explanatory diagram showing aliasing artifacts caused by the conventional method. (1>...Object A...RF pulse^1
1,^12...RF pulse for saturation B...Magnetic resonance signal B'...Aliasing artifact G1...Slice magnetic field G1...Phase encode magnetic field G1...Phase encode magnetic field for saturation G1...Signal Readout magnetic field L1: Imaging field of view LPI in the phase encoding direction, Ll
l! ... Saturated region S7... Step S8 of exciting the saturated region... Step S11 of signal collection... Step Sll... Step of reconstructing the tomographic image. In the drawings, the same reference numerals indicate the same or corresponding parts. . Figure 3 Figure 1 Figure 4? Display of the case Patent application No. 1-155573 Name of the invention Person who makes magnetic resonance imaging correction Relationship to the case Patent applicant address 2-2-3 Marunouchi, Chiyoda-ku, Tokyo Name (601) Mitsubishi Electric Corporation Representative Moriya Shiki

Claims (1)

[Claims] Applying an RF pulse and a gradient magnetic field consisting of a slice magnetic field for specifying a tomographic plane, a phase encoding magnetic field giving a phase encoding amount, and a signal readout magnetic field for frequency encoding to the subject, In magnetic resonance imaging, in which a signal collection sequence of receiving magnetic resonance signals from a subject is repeated multiple times while changing the amount of phase encoding to construct a desired tomographic image of the subject, before the signal collection sequence. , A saturation sequence is executed in which a saturation phase encoding magnetic field and a pair of saturation RF pulses having different transmission frequencies are applied to excite an area outside the field of view in the phase encoding direction to be imaged. magnetic resonance imaging.