JPH03131237A - Magnetic resonance imaging apparatus - Google Patents

Abstract

PURPOSE:To suppress artifacts so as to obtain an image free from blur without an extension of measurement time by correcting a difference of longitudinal magnetization caused by a longitudinal relaxation time before start of every projection by using a free induction damping signal. CONSTITUTION:A magnetic resonance imaging apparatus is equipped with a central processing unit(CPU) 1, a sequencer 2, a transmission system 3, a magnet 4 for generating static magnetic field, a reception system 5 and a signal processing system 6. Assuming that a magnitude of longitudinal magnetization before start of n-th projection is Mn-1, a free induction damping(FID) signal turns out to be a function of Mn-1 and a spin echo signal also a function of Mn-1. Therefore, this free induction damping(FID) signal is utilized for removing an influence of a longitudinal relaxation from the spin echo signal. It is thus possible to reduce artifacts due to a difference of longitudinal magnetization before start of every projection and to obtain a clear image free from blur.

Description

Detailed Description of the Invention [Field of Industrial Application] The present invention provides a nuclear magnetic resonance imaging apparatus that utilizes the nuclear magnetic resonance (NMR) phenomenon to obtain a tomographic image of a subject. Free induction damping (FID) is the difference in longitudinal magnetization at the start of each projection due to T1).
It is related to correcting signals and reducing artifacts.

(Prior Art) A magnetic resonance imaging apparatus (hereinafter referred to as an MRI apparatus) utilizes nuclear magnetic resonance phenomena to determine the density distribution of atomic nuclear spins (hereinafter simply referred to as spins) at a desired inspection site in a subject. It measures the relaxation time distribution, etc., and displays an image of the cross section of the subject based on the measured data.

This device is a static magnetic field generator that generates a static magnetic field of about 0.02 to 2 Tesla, as shown in Figure 1! A subject 7 is placed in the container 4. At this time, the spins in the subject perform precession with the direction of the static magnetic field as an axis at a frequency determined by the strength HO of the static magnetic field.

This frequency is called the Larmor frequency, Larmor frequency ν
0 is.

γ νo =Ho “−(
1) It can be expressed as 2π. Here, γ is the gyromagnetic ratio, which has a unique value for each type of atomic nucleus. If the angular velocity of Larmor precession is ω0, then because of the relationship ω0=2πνOφ...(2), ω0=γHo...(3
) is given by

Here, when a high frequency magnetic field (electromagnetic wave) with a frequency equal to the Larmor frequency ν0 of the atomic nucleus is applied by the high frequency irradiation coil 11, the spins are excited and transition to a high energy state. When this high-frequency magnetic field is cut off, the spins return to their original low energy state. The electromagnetic waves emitted at this time are called free induction decay (FID) signals, and the decay is the longitudinal induction time (Tl).

It is characterized by two time constants: transverse relaxation time (T2).

Here, we will discuss the second part of the pulse sequence spin echo method, which is currently commonly used in imaging with MHI equipment.
This will be explained using the diagram and FIG.

FIG. 2 schematically shows the pulse sequence of the spin echo method, and FIG. 3 schematically shows the behavior of macroscopic magnetization in a rotating coordinate system.

The atomic nucleus, which is bent in the static magnetic field, performs a differential motion at the Larmor frequency, and the macroscopic magnetization is directed in the direction of the static magnetic field (Z′).
(I).

In this pulse sequence, first, a 901 pulse 22 is applied, and then a 180° pulse 23 is applied after a time of Te/2, where the echo time is Te. 90'' pulse 2
After adding 2, each spin moves at a unique speed to X-Y
As each spin begins to rotate within the plane, a phase difference occurs between each spin over the course of one hour. A signal called the free induction damping signal 25 is generated here. Furthermore, when the 180' pulse 23 is applied, each spin is reversed symmetrically about the X axis as shown in FIG. 3 (IV), and the spins refocus at time Te to continue rotating at the same speed. form an echo signal.

Moreover, after the application of the 90° pulse 22, each spin tries to return to a thermal equilibrium state. This is called relaxation, and the progress of spin focusing on the component in the longitudinal (Z) direction is called longitudinal relaxation, and its time constant is called longitudinal relaxation time (T1).

The signals are measured as described above, but in order to construct a tomographic image, the spatial distribution of the signals must be determined, and a linear gradient magnetic field is used for this purpose.

A spatial magnetic field gradient is created by superimposing a gradient magnetic field on a uniform static magnetic field. This is used to give spatial location information to the signal. For this reason, the phase encoding gradient magnetic field 2 in FIG.
4 and a frequency gradient magnetic field 29 are used. A 5-phase encoded gradient magnetic field 24 using the pulse sequence as a basic unit.
for a constant repetition time (
TR) is repeated a predetermined number of times.

A desired tomographic image can be obtained by rearranging the measurement signals thus obtained in the order of phase encode gradient magnetic field strength and performing two-dimensional inverse Fourier transformation.

That is, the obtained signal 5(k-, ky) is ideally given by: Here, GP is the phase encoding gradient magnetic field strength, G is the frequency encoding gradient magnetic field strength considering the spin direction, 1. P represents the phase encoding gradient magnetic field application time, and P represents the frequency encoding gradient magnetic field application time.

Furthermore, ρ(x, y) can be expressed as...(5). Here, T1 is the longitudinal relaxation time, T2 is the transverse relaxation time, TR is the repetition time, and M is the static magnetic field direction component of the macroscopic magnetization vector before the start of projection (hereinafter referred to as longitudinal magnetization).

By performing a two-dimensional inverse Fourier transform on equation (4), the image A(x, y) can be expressed as A(x, y): (x, y) (6).

Here, k is an arbitrary constant, and M ideally does not change from projection to projection. That is, ideally, Mo = M! = M, z...Mll-1"・
= ”' (7). Here, M
n-1 represents longitudinal magnetization before the start of the n-th projection. However, in reality, the longitudinal magnetization before the start of projection differs for each projection due to the influence of the longitudinal relaxation phenomenon.

Further, Mn-1Mn-1= is established when the longitudinal relaxation time T1 is sufficiently shorter than the repetition time TR or when n is sufficiently large. That is, the above-mentioned artifacts can be suppressed by extending TR during imaging, or by excitation multiple times prior to measurement and starting measurement when n becomes sufficiently large.

Also, the imaging time T total is.

Ttotal=TR-PRJ・NSA+L-Tl;”
- given by (8). Here, PRJ is the number of projections, NSA is the number of integrations, and L is the number of times of excitation prior to measurement.

Regarding the above basic principles of MRI, please refer to Magnetic Resonance Imaging, (1988) (Magnet
ic Re5onance, I+Ilaging,
(1988)).

[Problem to be solved by the invention]

In the prior art, the magnitude of longitudinal magnetization before the start of each projection is strictly different, and the magnitude of the signal also affects the magnitude of longitudinal magnetization for each projection, resulting in blurring of the image.

In view of the problems of the prior art, it is an object of the present invention to suppress the artifacts without extending measurement time and obtain images without blur.

[Operation] According to the present invention, artifacts due to differences in longitudinal magnetization before each projection start can be reduced without extending measurement time, and clear images without blur can be obtained.

〔Example〕

Hereinafter, one embodiment of the present invention will be explained with reference to FIGS. 1 to 4.9 FIG. 1 is a block explanatory diagram of the overall configuration of a nuclear magnetic resonance imaging apparatus to which the present invention is applied, and FIG. 2 is a spin echo method. 3 is an explanatory diagram showing the macroscopic magnetization of the spin echo method in a rotating coordinate system. FIG. 4 is a schematic diagram of the pulse sequence of the spin echo method according to an embodiment of the present invention. A magnetic resonance imaging apparatus to which the present invention is applied will be explained with reference to FIG. 1, which is an explanatory diagram. This magnetic resonance imaging apparatus is roughly divided into a central processing unit (CPU) 1, a sequencer 2, a transmission system 3, a static magnetic field generating magnet 4, a reception system 5, and a signal processing system 6. do.

The central processing unit 11 (CPU) 3 executes the sequencer 2. Transmission system 3. Receiving system 5. It controls each of the signal processing systems 6. Sequencer 2
operates based on control instructions from the central processing unit 1,
A transmission system 3. transmits various commands necessary for data collection of tomographic images of the subject 7. Gradient magnetic field generation system 211 of static magnetic field generation magnet 4.
I am trying to send it to the receiving system 5.

The transmission system 3 includes a high-frequency generator 8, a modulator 9, and an irradiation coil 11 as a high-frequency coil.The modulator 9 modulates the amplitude of high-frequency pulses from the high-frequency oscillator 8 according to instructions from the sequencer 2. A predetermined pulsed electromagnetic wave is irradiated onto the subject 7 by amplifying the amplitude-modulated high-frequency pulse via a high-frequency amplifier and supplying it to the irradiation coil.

The static magnetic @ generating magnet 4 is for generating a uniform static magnetic field in any direction around the subject.

Inside this static magnetic field generating magnet, in addition to the irradiation coil]]-, gradient magnetic field coils 1 and 3 for generating gradient magnetic fields, and a receiving coil]-4 of the receiving system 5 are installed. 21- denotes a gradient magnetic field coil 13 having a configuration capable of applying a nine-independent gradient magnetic field in the directions of Cartesian magnetic probe axes that are orthogonal to each other; a gradient magnetic field power source 1-2 that supplies current to the gradient magnetic field coil; and a gradient magnetic field source 1-2. A sequencer 2 controls a power supply] and 2.

The receiving system 5 includes a receiving coil 14 as a high-frequency coil, an amplifier [5] connected to the receiving coil [4], a quadrature phase detector [6], and an A/D converter 17.
When the receiving coil 14 detects the MR signal, the signal 4a is sent to the amplifier 15. Quadrature phase detector 16. When converted into digital data via the A/D converter 17, the data is converted into two series of collected data sampled by the quadrature phase detector 16 at the timing according to the command from the sequencer 2, and then sent to the central processing unit 1. - I try to send it to.

The signal processing system 6 includes a magnetic disk 20. It has an external storage device such as an optical disk 19, and a display from a CRT 18, etc. When data from the receiving system 5 is input to the central processing unit, the central processing unit 1 performs signal processing, image reconstruction, etc. The process is executed, and the resulting desired cross-sectional image of the subject 7 is displayed as C.
It is displayed on the RT 18 and also recorded on the magnetic disk 20 of an external storage device.

Next, the change in longitudinal magnetization before the start of each projection will be explained using FIGS. 2 and 3.

Let Mo be the component of the spin in the direction of the static magnetic field (Z') in thermal equilibrium (I). First, the spins are excited by the 90° pulse 22 of the -th projection and are collapsed onto the X'-Y' plane. (11) From that point on, a longitudinal relaxation phenomenon occurs with a time constant of longitudinal relaxation time T1, and the longitudinal magnetization of the spin attempts to return to Mo (I?). Furthermore, it is inverted by the 180' pulse 23 (■).

It tries to return to Mo (■ to ■), but if TR is short or T1 is long, the second projection (B) starts before the longitudinal magnetization returns to Mo. The number of spins excited by the 90'' pulse 22 of the second projection is smaller than that of the first projection, reducing the signal.

In this way, the spins excited by the 90" pulse 22 gradually decrease after several projections.
It converges to a certain value, and the magnitude of the signal is also affected by it.

An embodiment in which the present invention is applied to a spin echo method will be described with reference to FIG. If the magnitude of longitudinal magnetization before the start of the n-th projection is Mo-1, free induction damping (F
ID) signal becomes a function of Mn-1. Furthermore, the spin echo signal also becomes a function of M7-1. For this reason. A free induction decay (FID) signal can be used reversibly to remove the influence of longitudinal relaxation from the spin echo signal.

As shown in Figure 4, the free induction damping (FID) signal 25
and spin echo signal 2G respectively.27.
28), and let them be A(t, n) and B(t, rl).

n represents the number of projections.

B(t,n) is a conventional spin echo signal.

The number of sampling points for A(n) is - Take one point per projection, or take multiple points (m points) per two projections at an arbitrary sampling time, average and smooth the points using the calculations shown below, and convert them to 1. It may also be point data.

Let Ai(n) be each of the plurality of data of A(n), and assume that m pieces of data are measured, then A(n):= (9) Or.

A(n)== ・(10) or, m , we obtain A(n).

C(t,n) to C(t,n)=-B(t,n)/A(n)=4
12). Here, k is an arbitrary constant. This may be done after the measurement of each projection is completed, at the same time as the echo signal measurement, or after all measurement paths Y.

Furthermore, 5(k-,ky)”C(k□p(n))...
・Set as (13). Here, p(n) is an arbitrary function representing the update order of the phase encoding gradient magnetic field. For example, when changing the phase encoding gradient magnetic field strength sequentially, p
Let (n)==n. By subjecting this 5(kX, ky) to a two-dimensional inverse Fourier transform, artifacts are suppressed and image power is improved.

Another case where the present invention is applied to the spin echo method is as shown in FIG.
(n), B(t, n), where 6n represents the number of projections. B(t,n) is a conventional spin echo signal. The number of sampling points of A(n) is ] points per projection, or 1 at any sampling time. A plurality of points (m points) may be taken for each projection, averaged by the following calculation, and then used as one point of data. Further, A may be measured for each projection or for any projection, and each of the plurality of data of A(n) is A i (n
) and m pieces of data are measured, then A(n)= ...(14) Or, Ai(n) A(n)= ...(15) Or, A(n) as obtain.

In the present invention, The relationship between A (n) and n is Approximately, By the minimum error square approximation method, α.

β.

Find the value of γ shall be.

or, The relationship between A (n) and n is A(n)==αn+β ...(19) By approximating the minimum error squared approximation, αt value of β In search of A' (n)=αn+β ...(20) It's fine.

C(t,n) to C(t,,n)=-B(t,n)/A'(n)-
(21).

Furthermore, 5(k-,ky)=C(k-9p(n))-
(22). Here, p(n) is an arbitrary function representing the update order of the one-phase encode gradient magnetic field. For example, when changing the phase encode gradient magnetic field strength sequentially, p(n)
n)=n. By subjecting this 5(kx, ky) to two-dimensional inverse Fourier transformation, an image with suppressed artifacts can be obtained.

Further, in the present invention, the measurement sequence is described using a spin echo, but it can also be performed with other sequences in total of $11.

〔Effect of the invention〕

According to the present invention, artifacts due to differences in longitudinal magnetization before each projection start can be suppressed, so that a clear tomographic image can be obtained.

[Brief explanation of the drawing]

Figure 1 is a block diagram of the overall configuration of a nuclear magnetic resonance imaging system to which the present invention is applied, Figure 2 is a schematic diagram of the pulse sequence of the spin echo method, and Figure 3 is the macroscopic magnetization of the spin echo method. An explanatory diagram showing this in a rotating coordinate system,
FIG. 4 is a schematic explanatory diagram of a pulse sequence of a spin echo method according to an embodiment of the present invention. ]... Central processing unit (CPU), 4... Static magnetic field generating magnet, 7... Subject, 24... Phase encoding gradient magnetic field, 25... Free induction decay (F I r)) signal ,
2G...Echo signal, 27...Free induction decay (F
ID) Signal measurement timing, 28...Echo signal measurement timing. Higure 3 0 Hitsuji 4

Claims (1)

[Claims] 1. Means for generating each magnetic field of a static magnetic field, a gradient magnetic field, and a high-frequency magnetic field, a means for detecting a nuclear magnetic resonance signal from an examination object, and a means for detecting a nuclear magnetic resonance signal from an examination object using the resonance signal. In a nuclear magnetic resonance imaging apparatus having means for reconstructing to obtain a composition-related image and means for displaying the obtained image, a means for correcting an echo signal using a free induction decay signal is provided. Features of magnetic resonance imaging equipment. 2. The nuclear magnetic resonance imaging apparatus according to claim 1, further comprising means for measuring the free induction attenuation signal for each projection and correcting the echo signal. 3. The magnetic resonance imaging apparatus according to claim 1, further comprising means for measuring the free induction attenuation signal using an arbitrary projection, performing arbitrary function approximation, and correcting the echo signal. 4. A claim characterized by providing means for measuring the free induction attenuation signal at a plurality of points for one projection, performing arbitrary averaging processing on the measurement points, and correcting the echo signal using one of the measured points. Magnetic resonance imaging device in Items 1 and 2. 5. Claims 1, 2, and 3, characterized in that means is provided for changing the phase encoding gradient magnetic field strength in an arbitrary order.
, 4 magnetic resonance imaging device. 6. The magnetic resonance imaging apparatus according to claim 1, further comprising means for correcting a difference in the static magnetic field direction component of the macroscopic magnetization before the start of each projection.