JPH0345244A - Magnetic resonance image picture method - Google Patents

Abstract

PURPOSE:To clearly image a tomographic image of a transfer part by suppressing the magnetic resonance signal generated from the static part of an inspected body and collecting the magnetic resonance signals generated from the transfer part, in emphasis. CONSTITUTION:Though the magnetic resonance signal B1 generated from a static part is suppressed by constituting each inclined magnetic field from a plurality of pulses, the magnitude of the magnetic resonance signal B2 is proportional to the volume occupied on an inspected body 1. Therefore, when the blood stream in a human body is image-displayed, the magnetic resonance signal B1 is obtained in the larger signal form in comparison with the signal B2, since the volume of the static part is larger than that of a transfer part (vein part). However, when the flow encode magnetic field in bipolar form in which the polarity is reversed in the sequence in two times is applied, and the magnetic resonance signal B is reduction-calculated, the magnetic resonance signal B1 supplied from the static part is suppressed. Therefore, the generation of the phase disturbance in the transfer part of the inspected body 1 is prevented, and the transfer part can be clearly image-pictured.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to magnetic resonance imaging for obtaining tomographic images using nuclear magnetic resonance (NMR), electron spin resonance (ESR), etc. This field relates to magnetic resonance imaging, which makes it possible to visualize the movement of blood and moving parts such as blood or lymph fluid.

[Prior Art] FIG. 6 is a block diagram showing a general magnetic resonance apparatus. In the figure, a subject (1) such as a human body is connected to a static magnetic field generator (2) that generates a static magnetic field in the Z-axis direction. ) has an examination table (3
) is located through.

A high frequency coil (4) that applies high frequency energy such as an RF (high frequency magnetic field) pulse A to the subject (1) and receives a magnetic resonance signal B from the subject (1) is connected to a matching device (
It is connected to the transmitter (7) and the receiver (8) via the transmitter (7) and the transmitter/receiver switching bag M (6).

Orthogonal to the object (1) 3 City 1fl (X Y Z
) direction, each gradient magnetic field coil (9>, (11), and (13) that applies the gradient magnetic field Gll, GP, and Gll is connected to each gradient magnetic field power source (work 0), (12), and (
4) respectively.

The sequence control device (15) includes an examination table (3), a transmitter (7), a receiver 1 (8), each gradient magnetic field power supply (1o), (1
2) and (14>), and the entire device is controlled in a predetermined sequence.

Generation of control data for image composition, processing of magnetic resonance signal B,
A computer (16) that performs image composition processing is connected to a receiver (8) and a sequence control device (15). Furthermore, an operation console (17) is connected to the computer (16) for inputting parameters necessary for image configuration, and the operation console (17) is equipped with an image display device for displaying tomographic images! (18
) are connected.

Next, a conventional magnetic resonance imaging method using a spin echo method will be described with reference to the pulse sequence diagram of FIG. 7.

Here, RF pulse 8 is a 90'' pulse, RF pulse A2 is 1806 pulses, magnetic resonance signal B is a spin echo signal, the gradient magnetic field in the X-axis direction is the signal readout magnetic field G for frequency encoding, and the gradient in the Y-axis direction is The magnetic field is a phase encode magnetic field Gp, the gradient magnetic field in the Z-axis direction is a slice magnetic field Gs for specifying a tomographic plane, and the case is shown in which an image is constructed by a two-dimensional Fourier transform method based on a magnetic resonance signal B.

First, the subject (1) is connected to the static magnetic field generator (2), the high frequency coil (4), each gradient magnetic field coil (9), (11) and (
13) and drive the high frequency coil (4) and Z-axis gradient magnetic field coil (13) to simultaneously apply a 90″ pulse^1 and a slice magnetic field of 0.1 to the subject (1). As a result , energy is supplied to the nuclear spins in the desired @layer plane in the object (1), and the center position N of the 90″ pulse ^1
(at the peak point), the phase of the nuclear spins begins to be disturbed (dephased).

Next, drive the X-axis gradient magnetic field coil (9) and the Y-axis gradient magnetic field coil (11) to apply a signal readout magnetic field of 0.1 and a phase encode magnetic field GP, and then apply 2 and a slice magnetic field G to the 18° pulse. , 2 are applied. At this time, in order to align (rephase) the phase of the spins in the slice direction when the slice magnetic field G51 is applied, the center position of the slice magnetic field G,2 is shifted from the center position of the 180'' pulse 82.

After that, a signal readout magnetic field G112 with the same polarity as G111
While applying the magnetic resonance signal B, the computer (1
6) Take it inside. At this time, the signal readout magnetic field GR2
The point in time when the pulse area of matches the pulse area of GPL, that is, the point in time when echo time TE has elapsed from the center position of the 90° pulse, is the center position (peak value) of the magnetic resonance signal B.
becomes. Therefore, the timing at which the magnetic resonance signal B reaches its peak value is determined by the 90° pulse ^1 and the signal readout magnetic field GR.
2 depends on the application timing. Further, data of the magnetic resonance signal B is collected at a predetermined number of sampling points during application of a signal readout magnetic field of 0.2.

The above sequence is repeated a number of times corresponding to a predetermined number of pixels N (for example, 256) while changing the phase encode amount Ki determined by the pulse area of the phase encode magnetic field G2 at a predetermined pitch (see broken line). For example, , when the number of pixels of the tomographic image is NXN, the number of sampling points for one magnetic resonance signal B is N or more, and the number of times the signal is collected is N times.Then, the computer (I6) performs two-dimensional Fourier transform on the pulse train of the magnetic resonance signal B to reconstruct a tomographic image of a desired matrix size NXN, and displays it on an image display device (18).

Here, the signal readout magnetic field GRI was applied between the RF' pulses 81 and 82, but the signal readout magnetic field 0.
2 may be applied with opposite polarity. Alternatively, by aligning the center positions of the slice magnetic field Gs2 and the 180° pulse 82, a slice magnetic field (shaded area) for aligning the spins in the slice direction may be applied with opposite polarity immediately after the slice magnetic field GI11 is applied. .

Next, a conventional magnetic resonance imaging method using a gradient field echo method will be described with reference to the pulse sequence diagram of FIG. In this case, the RF pulse A has a flip angle of α6 (for example, 90°
1806- pulse is not applied. Moreover, the magnetic resonance signal B is a gradient field echo signal.

First, apply an RF pulse A and a slice magnetic field Gg1,
Subsequently, a slice magnetic field G81' having the same pulse area as the shaded portion of the slice magnetic field Gg1 and opposite polarity is applied to align the spin phases, and a signal readout magnetic field G111' and a phase encode magnetic field GP are applied.

Thereafter, the magnetic resonance signal B is received while applying a signal readout magnetic field of 0.2 with the polarity reversed, and is taken into the computer (16) in the same manner as described above, and an image is constructed by two-dimensional Fourier transformation.

[Problem to be solved by the invention] The conventional magnetic resonance imaging method is based on the subject (1) as shown below.
Each gradient magnetic field G8, G, l and G
The pulse waveform of p is determined, and the magnetic resonance signal B is received. For this reason, the pulse sequence is determined so that the phase of the nuclear spins constituting the stationary part of the object (1) becomes coherent, and
The phase of the nuclear spin becomes incoherent, and the intensity of the magnetic resonance signal B from the moving part is significantly suppressed. Therefore, even when trying to diagnose blood flow in blood vessels, it has been difficult to create a tomographic image of the moving portion of the subject (1).

This invention was made to solve the above-mentioned problems, and it significantly suppresses the magnetic resonance signals generated from the stationary part of the subject, emphasizes and collects the magnetic resonance signals generated from the moving part, and collects the magnetic resonance signals generated from the moving part. The purpose of this study is to obtain a magnetic resonance imaging method that can produce clear tomographic images of a region.

[Means for Solving the Problems] The magnetic resonance imaging method according to the present invention includes: a slice magnetic field applied in response to an RF pulse; a phase encode magnetic field;
At least one of the signal readout magnetic fields applied in response to the magnetic resonance signal is composed of a plurality of pulses, and the pulse area of each gradient magnetic field is determined by the amount of phase change in the spins of the stationary part and the moving part of the object. is set to be incoherent with respect to the stationary part and coherent with respect to the moving part, thereby suppressing the magnetic resonance signal from the stationary part, and at the same time suppressing the magnetic resonance signal from the stationary part.
a first sequence in which a flow-encoding magnetic field is applied to shift the phase of the spins in the moving part from the phase of the spins in the stationary part in a first direction without disturbing the phase of the spins in the moving part, and a flow-encoding field having an opposite polarity with respect to the first flow-encoding field; applying a second flow encoding magnetic field to shift the phase of the spins of the stationary part in a second direction from the phase of the spins of the stationary part without disturbing the phase of the spins of the moving part; By taking the difference between the magnetic resonance signals obtained in the sequence, only the magnetic resonance signals from the moving part are emphasized and collected.

Further, in the magnetic resonance imaging method according to another aspect of the present invention, a bipolar first flow-encoding magnetic field is applied in the direction of motion of a moving part of the subject, so that the moving part remains stationary without disturbing the phase of the spins. A first sequence that shifts the phase of the spins of the part to shift 1 in a first direction, and applying a second flow-encoding magnetic field of opposite polarity to the first flow-encoding magnetic field to shift the phase of the spins of the moving part. A second sequence in which the phase of the spins of the stationary part is shifted in a second direction without disturbing the spins is carried out, and the difference between the magnetic resonance signals obtained in the first and second sequences is taken, and the phase of the spins of the moving part is shifted from the moving part. At the same time, a gradient magnetic field for dephasing in the same direction as the flow encoding magnetic field is applied at a timing consecutive to the application sequence of the first flow encoding magnetic field, and the spins in the stationary part are collected. This is designed to disturb the phase.

[Operation] In the present invention, a plurality of gradient magnetic field pulses are used to disturb the phase of the spins in the stationary portion, thereby making the spins coherent by correcting the phase disturbance of the spins in the non-coherent I and moving portions. In addition, by applying bipolar first and second flow encoding magnetic fields whose polarities are reversed for each sequence, the phase of the spins in the stationary part can be shifted in both directions without disturbing the phase of the spins in the moving part. 1 ~ and the difference between the magnetic resonance signals obtained in two sequences =
1, only the magnetic resonance signal from the moving part is emphasized and clearly separated from the magnetic resonance signal from the stationary part.

In another aspect of the present invention, bipolar first and second flow-encoding magnetic fields enhance and clearly separate the magnetic resonance signals of the moving portion from the stationary portion, and the first and second flow-encoding magnetic fields - A dephasing gradient magnetic field applied continuously with the magnetic field disturbs the phase of the stationary part and suppresses the magnetic resonance signal from the stationary part.

[Example] Hereinafter, an example of the present invention will be described with reference to the drawings. still,
The configuration of the magnetic resonance apparatus to which this invention is applied is as shown in FIG. 6, and it is only necessary that the contents of the sequence control bag ff (15) and part of the calculation program in the computer (16) be changed. .

FIG. 1 is a pulse sequence diagram showing an embodiment of the present invention when the two-dimensional Fourier transform method and gradient field echo method are used similarly to FIG. 8.

2- In FIG. 1, the slice magnetic field G, ]1-1 and the signal readout magnetic field G R21 are the slice magnetic field G81 in FIG.
' and a signal readout magnetic field of 0.2, but the pulse areas of each gradient magnetic field G sll and G R21 are different from those described above. Further, the phase encode magnetic field Gp is composed of a pair of gradient magnetic fields GP1 and Gp2 whose polarities are reversed.

Furthermore, before the phase encoding field GP is applied, a first flow encoding field G, 11 and G, 12 of bipolar type with reversed polarity is applied, and in the following sequence a bipolar type with reversed polarity is applied. The second flow encodes magnetic fields G, 21 and G, 22 (see dashed lines)
is applied. These flow-encoding magnetic fields G1 are applied in the direction of motion of the moving part of the subject (1), ie in the direction of motion of interest for imaging purposes.

First, conditions for determining the pulse area when dividing each gradient magnetic field G, , Gp, and GR into a plurality of pulses will be explained. Generally, when a gradient magnetic field G (t) whose intensity changes over time is applied, a position P in the direction of this gradient magnetic field G (t)
The amount of phase change φ of the nuclear spin of the specimen (1) at (t) is expressed by the equation integrated over time, φ−γS G (t) P (t) dt ・・
・■ However, γ is expressed as the nuclear gyromagnetic ratio. Here, assuming that the nuclear spin position p (t) of the object (1) changes over time, the stationary part is zero-order motion, the moving part at a constant speed is first-order motion, and the moving part with constant acceleration is If we assume that the moving part has a second-order motion and the higher-order moving part has an i-order motion, then the position P (t) of the nuclear spin is determined by the motion order i (-
0, 1, 2, ...), P (t) - Σact' ... ■ However -
at: Can be expressed as a coefficient that gives the magnitude of the first-order motion.

It can be seen that by expressing the nuclear spin position of the object (1) as shown in equation (2) and substituting this into equation (2), the amount of phase change φ can be arbitrarily set according to the motion order i of the target spin. In other words, in principle, if the gradient magnetic field G (t) is divided into multiple pulses and the area of each pulse is set appropriately, only the movement of the moving part of interest can be extracted and the amount of phase change φ for that movement can be kept constant. (or zero).

FIG. 1 shows a sequence in which magnetic resonance signals B from a moving part of a subject (1) moving in an arbitrary direction at a constant speed <1=1 can be collected coherently without causing any phase disturbance. Here, we will show a case in which a moving part that moves at a constant speed in all gradient magnetic field directions is to be imaged, but if the direction of the moving part can be specified in advance, the gradient magnetic field in only that direction of motion can be divided into multiple pulses. do it.

First, the pulse area setting conditions of the slice magnetic field G8 applied in response to the RF pulse A will be explained. If a plurality of temporally changing slice magnetic fields G s 1 and G, IN (shaded area) are G, (t), then the slice direction position pg (
The amount of phase change φ8 of the nuclear spin of t) is, from the formula (■), R
Slice magnetic field GI from the peak point (center) of F pulse A
! Integral formula until the end of application of (t), φ8−γ5 G,
(t) P g(t)dt. Therefore, the ratio of the pulse areas 3.1 and 5l111 of each diagonal part of the slice magnetic fields G,1 and G511 is set for each pulse so that the phase change amount φ8 of the moving part moving at a constant speed 15 in the slice direction becomes O. Considering the polarity of , it is set as S,1:S,11=3+-1...■. Here, the slice magnetic fields GI11 and G1111
Since the application time of each shaded portion is constant and each pulse waveform is considered to be rectangular, the relationship of equation (2) corresponds to the pulse intensity (wave height) of each slice magnetic field G81 and G, 11.

Similarly, for a moving part that moves at a constant speed in the phase encoding direction, the phase change amount φ2 is set constant. That is, if the time-varying phase encoding magnetic field G pll and G, 12 are C;P(t>), the amount of phase change φ2 of the nuclear spin at the position PP(t) in the phase encoding direction is
The integral equation for the application time of the phase encoding magnetic field GP(t) is given by φ2=γS G p(t) P p(t)dt. Therefore, in order to keep the phase change amount φ2 constant, the ratio of the 6 pulse areas S pll and S 12 of each of the phase encode magnetic fields GP11 and G p12 is S ,1
1: S P12=-↑:1...Set as ■. Thereby, the phase encoding magnetic field Gp is applied to the nuclear spins of the moving part that moves at a constant speed in the phase encoding direction so that the phase change amount φ2 becomes a predetermined value. It goes without saying that this phase encode amount changes at a predetermined pitch as indicated by the broken line for each signal acquisition sequence, as described above.

Next, regarding the signal readout magnetic field G8, the pulse area is set so that the phase change amount φ8 becomes O for the moving part that moves at a constant speed in the direction of the signal readout magnetic field. That is, the phase change amount φ8 of the nuclear spin at the signal readout magnetic field direction position p+t(t) is φ8−γS a R(t) P
, (t)dt, and φ8-O, the ratio of the pulse areas SR1 and SR21 of each diagonal shaded part of the signal readout magnetic fields GP1 and G, 121 is Spl: 5R21--3: 1... ■It is set to.

By the gradient magnetic field sequence that does not satisfy the above equations 1 to 3, the magnetic resonance signal B from the moving part moving at a constant speed in each gradient magnetic field direction is acquired without any phase hole, and the magnetic resonance signal B from the stationary part is obtained. Since the movement is suppressed, a somewhat clear tomographic image of the moving part can be obtained.

However, in the present invention, a bipolar flow-encoding magnetic field G,
By applying 11 and G 2 , 1.2 (G 2 , 21 and G 2 , 22), the magnetic resonance signal originating from the moving part is actively phase-shifted from the magnetic resonance signal originating from the stationary part. That is, the phase of the nuclear spins in the moving part is shifted from the phase of the nuclear spins in the stationary part without disturbing the phase of the nuclear spins in the moving part. Here, the application timing of the flow encode magnetic field G is set before the application of the phase encode magnetic field GP, but it may be set at any timing as long as it is before the application of the signal readout magnetic field GR.

FIG. 2 is an explanatory diagram showing the phase relationship between the magnetic resonance and one signal B1 and B2 from the stationary part and the moving part of the object,
Here, the coordinates XYZ in FIG. 6 are shown converted into rotational coordinates X'Y'Z' corresponding to the nuclear spin.

In Fig. 2, by applying the RF pulse A (α° pulse), the macroscopic magnetization M of the nuclear spin changes to α with respect to the Z′ axis.
Only ° has fallen. Also, magnetic resonance signals B1 and B from the stationary part and the moving part shown projected onto the X'Y' plane.
2 (corresponding to the phase of each nuclear spin) points in the Y'-axis direction before applying the flow encoding magnetic field G. However, in the first sequence, by applying the flow encoding magnetic fields G111, G, and 12 of the first step, the magnetic resonance signal B2 of only the moving part changes to a phase change amount φ
, l.

In general, the time-varying flow-encoding magnetic field G, (t
) is applied, the flow encoding direction position p, (t)
The amount of phase change of the nuclear spin φ, is φ, −γS G
t(t)Pr(t)dt...It is represented by ■. Here, if we consider the case where a flow encode magnetic field G, (t) is set for a moving part that moves at a constant speed ■ and a constant acceleration a in the flow encode direction, the flow encode direction value Hp, (t) is set. The nuclear spin is given by equation (2): Pf(t)=alt+a2t2 where al: coefficient of velocity component B2: coefficient of acceleration component. Substituting this into Equation 0, the amount of phase change φ due to the flow encode magnetic field G is φ, −γ5 a
, (t) dt −7S t(v tT + a T 2) −■
becomes. However, V t = a 1. a = a2.

Therefore, the flow encoding magnetic field G, so as to satisfy the equation 0,
By setting the pulse area S and the application time T, it is possible to give a desired phase change amount φ to the moving part.

Here, the first signal applied in the first signal acquisition sequence is
The ratio of the respective pulse areas S, 11 and S, 12 of the flow encoding magnetic fields G, 11 and G, 12 is as shown by the solid line in Fig. 1, S, 11: S, 12 = 1 + -1...・■Set to 0. As a result, as shown in FIG.
φ, changes by 1 from φ.

Subsequently, in a second signal acquisition sequence, a second flow-encoding magnetic field G, zt and G of opposite polarity to the first flow-encoding magnetic fields G, 11 and G, 12 is added, without changing the other gradient magnetic field sequences. , 22 (see broken line) is applied. At this time, the magnetic resonance signal B2 from the moving part is
As shown in the figure, the phase changes by φ,2 in the opposite direction to that in FIG. 2, and the phase of the magnetic resonance signal B1 from the stationary part is Y'
It remains axial. This is equivalent to the effect that occurs when the direction of movement of the moving part is reversed.

Through the above two signal acquisition sequences, imaging for one phase encode amount is completed. A computer (16) subtracts the magnetic resonance signal B obtained from the two imaging operations, and uses the subtracted magnetic resonance signal B as image data to be used for subsequent imaging by Fourier transform. At this time, even if the data obtained after Fourier transformation is subtracted, the same result can be obtained in principle.

The magnetic resonance signal B2 from the moving part is given a phase change amount φ,l as shown in FIG. 2 by applying the first flow encoding magnetic field Grll and G,12 in the forward direction.
Furthermore, by applying the second flow encode magnetic fields G, 21 and G, 22 in the opposite direction, a phase change amount φ, 2 in the opposite direction is given as shown in FIG. The signal strength is almost doubled.

On the other hand, since the magnetic resonance signal B1 from the stationary part does not change phase no matter which flow encoding magnetic field is applied,
The signal strength becomes almost 0 by the subtraction process. Therefore, the tomographic image of the moving part finally obtained by Fourier transformation is not affected by the nuclear spin of the stationary part and becomes extremely clear.

The above is an example of imaging a tomographic image of a moving part that moves at a constant speed. Next, a case will be described where a tomographic image of a moving part that moves at a constant speed and constant acceleration is created.

FIG. 4 shows a sequence in which magnetic resonance signals from a moving part of the subject (1) moving at a constant speed and constant acceleration in any direction can be collected coherently without causing any phase disturbance. In this case, the shaded portions of each gradient magnetic field G8 and G, and the phase encode magnetic field GP are each divided into three pulses.

First, regarding the moving part in the direction of the slice magnetic field, the amount of phase change φ of the nuclear spin due to the slice magnetic field a5(t)
8 as O, the slice magnetic field G, 1. Pulse area Sn], S s12 of each shaded area of G, 1.2 and G□13
and SIl+13 are set as follows: S, 1:5I112:S, 13=(1,1+δ)knee(7+2δ):(2+δ)...■. However, δ is each slice magnetic field G, 1 and G,
This is a correction amount related to the application time and rise time of No. 11.

Regarding the moving portion in the direction of the phase encode magnetic field, in order to keep the phase change amount φ2 due to the phase encode magnetic field ap(t) constant, each phase encode magnetic field G pl3, G
Pulse area of pl, 4 and GP15 S pl3, S P
The ratio of 1423 and S pl5 is 5p13:5p14:5p15''(-2+δ'):(
3-2δ'):(-1+δ')-...[Phase] Set. However, δ' is a correction amount related to the magnitude of the constant velocity component and the constant acceleration component, the application time and rise time of each phase encode magnetic field, and the like.

Furthermore, regarding the moving part in the direction of the signal readout magnetic field, in order to make the amount of phase change φR due to the signal readout magnetic field C;+t(t) 0, the signal readout magnetic field G , 1, G
and the pulse area of each shaded part of G R22 SR1, S, 1
The ratio of 12 and 5822 is S, 1:S, 12:S, 22--(11+δ): (
7+2δ) Knee (2+δ)...■ Set.

With each gradient magnetic field sequence that satisfies the above 0-0 formula,
It is possible to acquire magnetic resonance signals from a moving part that moves at a constant speed and -constant acceleration in the direction of the slice magnetic field, the phase encode magnetic field, and the signal readout magnetic field without phase disturbance, and the magnetic resonance signals from the stationary part. Since the movement is suppressed, a tomographic image of the moving part with a certain degree of clarity can be obtained.

Similarly to the above, in this invention, a flow encoding magnetic field Gf is applied to a moving part in an interesting direction to be imaged, and a tomographic image of the moving part moving at a constant speed and constant acceleration in the direction of the flow encoding magnetic field is obtained. Make it even clearer.

In this case as well, from equation (■), each flow encode magnetic field a, 1
The pulse area ratio of 1 and G, 1.2 is set to 1 - 1, as in equation (2). Also, the first flow encoding magnetic field G,
11 and G, 12 and the second flow encoding magnetic field G
, 21 and G, 22, the phase of the nuclear spin of the moving part is shifted as shown by φf and φt2 in FIGS. 2 and 3. Therefore, by subtracting the magnetic resonance signals B obtained in two sequences, it is possible to emphasize and separate the magnetic resonance signal B2 of only the moving portion.

In this way, imaging is performed for motion orders up to second order, but similarly, for moving parts having motion components of third order or higher, multiple pulse areas of each gradient magnetic field are used to compensate for the amount of phase change. can be set. At this time, phase disturbance compensation can be performed in any required gradient magnetic field direction depending on the moving part of the subject (1). or,
For any flow-encoding magnetic field direction, magnetic resonance signals from stationary parts can be suppressed and magnetic resonance signals from moving parts can be enhanced.

That is, by configuring each gradient magnetic field with a plurality of pulses, the magnetic resonance signal B1 generated from the stationary part is suppressed, but the magnitude of the magnetic resonance signal B is proportional to the volume occupied by the subject (1). It has been known. Therefore, for example, when imaging blood flow in a human body, the volume of the stationary part is larger than that of the moving part (blood vessel part), so the magnetic resonance signal B1 is different from B2.
A very large signal is obtained compared to the However, as in the present invention, applying a bipolar flow-encoding magnetic field whose polarity is reversed in two sequences,
By subtracting the magnetic resonance signal B, the magnetic resonance signal B1 from the stationary part is further suppressed. Thereby, the moving portion of the subject (1) can be clearly imaged without causing any phase disturbance of the moving portion.

Next, another embodiment of this invention will be described.

FIG. 5 is a pulse sequence diagram showing another embodiment of the present invention, and the gradient magnetic field sequence in three directions is shown in the eighth embodiment.
As shown in the figure (as before), the first flow encoding magnetic field G
, 11 and G, 12, a dephasing gradient magnetic field G0 is applied in the same direction as the flow encoding magnetic field direction.

In this case, as described above, by applying the first and second flow-encoding magnetic fields of opposite polarity to each other in two signal acquisition sequences and taking the difference between the magnetic resonance signals B obtained in each sequence, Magnetic resonance signal B1 from part
suppress.

At this time, the first flow encoding magnetic field G, 11 and G,
12, by further applying a dephasing gradient magnetic field G ta, the magnetic resonance signal B1 from the stationary part is
is further suppressed, and the tomographic image of the moving part becomes clearer.

27- Here, the application timing of the dephasing gradient magnetic field G r was set immediately after the application of the first flow encoding magnetic field, but if the timing is continuous with the first flow encoding magnetic field, It may be set immediately before applying the encoding magnetic field. Furthermore, if the strength of the first flow encoding magnetic field is set appropriately, a pair of flow encoding magnetic fields G,
It may also be inserted in the middle of 11, G, and 12. Further, the intensities of the dephasing gradient magnetic fields G and d are amounts that can be adjusted and set as appropriate depending on the object to be imaged, and may be pulses of not only positive polarity but also negative polarity.

First, the bipolar first and second flow encoding magnetic fields change the phase of the nuclear spins in the moving part, and the magnetic resonance signal B2 changes in phase change amount φ1. and φ, shifted by 2. In addition, due to the dephasing gradient magnetic field G td,
The phase of the nuclear spins of the moving part moving in the flow encoding direction is disturbed.

At this time, the volume ratio within the subject (1) in the flow encoding direction is that the volume occupied by other stationary parts is much larger than that of the blood vessel to be imaged. Therefore, the phase of nuclear spins in moving parts such as blood flow is hardly disturbed, and the phase only in stationary parts is greatly disturbed. As a result, the magnetic resonance signal B1 from the stationary portion is further suppressed, and the tomographic image of the moving portion can be of high quality.

In the second sequence, the second flow encode magnetic fields G, 21 and G, 22 of opposite polarity are applied as shown by the broken line, but in the first sequence described above, the phase of the nuclear spins in the stationary part is already disturbed. Therefore, there is no need to apply a gradient magnetic field for dephasing.

Here, the pulse area ratio of the flow encoding magnetic fields G, 11 and Gd2 is set to 1, so as mentioned above,
A tomographic image of a moving part at a constant speed and constant acceleration is clearly imaged.

As shown in FIG. 5, the method of applying the dephasing gradient magnetic field G ta is particularly effective when applied to imaging blood flow in the human body because the pulse sequence can be set relatively easily.

In each of the above embodiments, the case where the pulse waveform of each gradient magnetic field is trapezoidal is shown, but even with other pulse waveforms such as a sine wave, as long as the ratio of each pulse area satisfies the above relationship. good.

Further, although the case where the 1tIr layer image is obtained from the magnetic resonance signal B using the two-dimensional Fourier transform method has been shown, a three-dimensional Fourier transform method may also be used.

Further, although the case where the signal acquisition sequence is the gradient I-echo method has been described, the present invention can also be applied to a spin echo method as shown in FIG.

In this case, the slice magnetic field GI]2 corresponding to the 180° pulse ^2 may also be divided into a plurality of pulses as shown in the shaded area in FIG. 1 or FIG.

Furthermore, the collected magnetic resonance signal B is not limited to a gradient echo signal, but may be another magnetic resonance signal, for example, an NMR signal such as a spin echo signal, or an electron spin signal. In the case of an electron spin signal, the nuclear gyromagnetic ratio γ, which provides the amount of phase change, represents the electron spin rotation ratio.

"Effects of the Invention" As described in (2) below, according to the present invention, a slicing magnetic field applied in response to an RF pulse, a phase encoding magnetic field,
At least one of the signal readout magnetic fields applied in response to the magnetic resonance signal is composed of a plurality of pulses, and the phase change amount of the spins of the stationary part and the moving part of the object is incoherent with respect to the stationary part. The area of each pulse is set so as to be coherent with respect to the moving part, suppressing the magnetic resonance signal from the stationary part, and
a first sequence of applying a bipolar first flow-encoding magnetic field in the direction of motion of the moving part to shift the phase of the spins of the moving part in a first direction from the phase of the spins of the stationary part; applying a second flow-encoding magnetic field of opposite polarity to the flow-encoding magnetic field to shift the phase of the spins in the moving part in a second direction from the phase of the spins in the stationary part; , the difference between the magnetic resonance signals obtained in the first and second sequences is taken, and only the magnetic resonance signals from the moving part are emphasized and collected, which improves the image quality of the tomographic image of the moving part. The achieved magnetic resonance imaging method has three effects.

According to another aspect of the present invention, a bipolar first flow encoding magnetic field is applied in the direction of motion of the moving part of the object, so that the phase of the spins of the moving part is made to be different from the phase of the spins of the stationary part. a first sequence that shifts I in one direction and a second sequence of opposite polarity with respect to the first flow-encoding magnetic field.
a second sequence in which a flow-encoding magnetic field is applied to shift the phase of the spins in the moving part from that in the stationary part in a second direction; By taking the difference in the magnetic resonance signals, only the magnetic resonance signals from the moving part are emphasized and collected.
A dephasing gradient magnetic field in the same direction as the flow encoding magnetic field was applied at a timing consecutive to the application sequence of the first flow encoding magnetic field to disturb the phase of the spins in the stationary part so that they did not contribute to the magnetic resonance signal. This has the effect of providing magnetic resonance imaging that achieves high-quality tomographic images of moving parts with a simple sequence.

[Brief explanation of drawings]

2- FIG. 1 is a pulse sequence diagram showing one embodiment of the present invention, FIG. 2 is an explanatory diagram showing the amount of phase change of the magnetic resonance signal from the moving part when the first flow encoding magnetic field is applied, and FIG. Fig. 3 is an explanatory diagram showing the amount of phase change of the magnetic resonance signal from the moving part when the second flow encoding magnetic field is applied, Fig. 4 is a pulse sequence diagram showing another embodiment of the present invention, and Fig. 5 is a pulse sequence diagram showing an embodiment of another invention of the present invention, FIG. 6 is a block diagram showing a general magnetic resonance apparatus, and FIG. 7 is a diagram showing a conventional magnetic resonance imaging method using a spin echo method. Pulse Sequence Diagram FIG. 8 is a pulse sequence diagram that does not show conventional magnetic resonance imaging using the gradient field echo method. (1)...Object A...RF pulse Gs...
...Slice magnetic field GR...Signal readout magnetic field GP
. . . Phase encoding magnetic field B . . . Magnetic resonance signal G , 11, G , 12 . . . First flow encoding magnetic field G , 21. G, 22...Second flow encode magnetic field Gz...Gradient magnetic field for dephasing Roto...Magnetic resonance signal from the stationary part B2...Magnetic resonance signal φt1, φ12 from the moving part
...Amount of phase change In the figures, the same reference numerals indicate the same or equivalent parts.

Claims (2)

[Claims] (1) A sequence in which an RF pulse, a slicing magnetic field, a phase encoding magnetic field, and a signal readout magnetic field are applied to a subject, and a magnetic resonance signal from the subject is collected, and the amount of phase encoding by the phase encoding magnetic field is collected as a signal. In a magnetic resonance imaging method in which a tomographic image of the subject is imaged based on the acquired magnetic resonance signals, the slice magnetic field is applied in response to the RF pulse; At least one of the phase encoding magnetic field and the signal readout magnetic field applied in response to the magnetic resonance signal is composed of a plurality of pulses, and the pulse area of the plurality of gradient magnetic fields is determined by the pulse area of the plurality of gradient magnetic fields. The amount of phase change of the spins is set to be incoherent with respect to the stationary part and coherent with respect to the moving part, thereby suppressing the magnetic resonance signal from the stationary part, and suppressing the magnetic resonance signal from the moving part. a first sequence of applying a first flow-encoding magnetic field bipolar in the direction of motion to shift the spins of the moving part in a first direction relative to the spins of the stationary part without disturbing the phase of the spins; applying a second flow-encoding magnetic field of opposite polarity to the first flow-encoding magnetic field to shift the spins of the stationary portion in a second direction without disrupting the phase of the spins of the moving portion; 2 sequences, and the difference between the magnetic resonance signals obtained in the first and second sequences is taken to emphasize and collect only the magnetic resonance signals from the moving part. magnetic resonance imaging. (2) A sequence in which an RF pulse, a slicing magnetic field, a phase encoding magnetic field, and a signal readout magnetic field are applied to a subject, and a magnetic resonance signal is collected from the subject, and the amount of phase encoding by the phase encoding magnetic field is collected as a signal. In a magnetic resonance imaging method in which a tomographic image of the subject is imaged based on the acquired magnetic resonance signals while changing the order of the magnetic resonance signals, a bipolar first
A flow-encoding magnetic field is applied to the spins of the stationary part without disturbing the phase of the spins of the moving part.
and applying a second flow-encoding magnetic field of opposite polarity to the first flow-encoding magnetic field to shift the motion of the stationary portion without disturbing the phase of the spins of the moving portion. a second sequence in which the spins are shifted in a second direction; and the difference between the magnetic resonance signals obtained in the first and second sequences is calculated to determine the magnetic resonance from the moving part. In addition to emphasizing and collecting only the signals, applying a gradient magnetic field for dephasing in the same direction as the flow encoding magnetic field at a timing that is continuous with the application sequence of the first flow encoding magnetic field, and dephasing the spins of the stationary portion. A magnetic resonance imaging method characterized by phase disturbance.