JPS63130056A - Partial selective nuclear magnetic resonance data lead-out method - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] This invention relates to a method for obtaining nuclear magnetic resonance information within a selected portion of a subject.

[Conventional technology]

Conventional techniques for obtaining nuclear magnetic resonance information within selected portions of a subject include the VSE method of W, P, At1E et al. (J, Magn, Re5on, +56, 350
-354 (1984), Japanese Patent Application Laid-open No. 59-645)
1. ■R, J, 0RDIDGE et al.'s l5IS method (J, M
agn, Re5on, 66.283-294 (19
86),).

■5PAR3 method of P, R, LIIYTEN et al. (J,
Magn, Re5on,.

Ken, 14B-155 (1986), ) and ■P, A,
BOTTOML! There is a method of implementing localized NMR chemical shift spectroscopy of J (Japanese Patent Application Laid-open No. 107246/1983).

The pulse sequence in the VSE method (①) is as shown in Figure 10, with a non-selected 90° pulse and a selected 4 pulse sandwiching it.
A slice in one direction is selected by a set of 5° +45° pulses (i.e., a non-selective 90° pulse is inserted in the center of a selected 90° pulse), and the inner spins have a component in the Z or -Z direction. , but the outside is boiled. In other words, a sharp 90° pulse is applied to the subject while applying a static magnetic field in the Z-axis direction and an X-axis gradient magnetic field Gx in the same direction but whose intensity gradually changes along the X-axis. When applied to the specimen, this 90° pulse has multiple frequency components, so specific spins are excited in any part of the specimen on the X axis, that is, non-selective excitation is performed. Since there is almost only a specific frequency component, only the part (selected part) cut at a specific position on the From this state, the spins are inverted by 180° (in the Z-axis or -Z-axis direction) and become unexcited RM.On the other hand, since the gradient magnetic field GK is applied for a relatively long time, it The excited spins (spins tilted by 90 degrees) rotate according to the strength of G at that position, that is, the spin axis rotates, and each spin rotates the non-selected part separately, resulting in so-called spoiling. The combined value of the free induced signals becomes zero as a whole.

If the same thing is done for the Y-axis gradient magnetic field and the Z-axis gradient magnetic field, the spin direction will be the Y-axis or -Z only at the intersection (selected portion) of the cross section perpendicular to each of the This will become the axis, and all spins in other non-selected parts will be spoiled. After this, by applying a 90' excitation pulse, a free induction signal (hereinafter FID) from the selected part is generated.
signal) is observed.

The pulse sequence in the 15IS method (2) is as shown in FIG. In this case, the partial selection sequence in the VSE of (2) has been changed to a 180° pulse for selection of the x, y, and z axes. In other words, while applying the X-axis gradient magnetic field G, 18
Give a 0° pulse to tilt a specific spin in a vertical section at a selected position on the X-axis = 180 degrees in the Z-axis direction), Y
After doing the same for the axis and Y axis, a 90° excitation pulse is applied. However, as shown in Figure 12, the sequence is performed eight times, and the OFF is <18 without reversing the spin.
0'' pulse is not applied), ON indicates spin reversal, and the FID signal obtained in this 8-time sequence is
Addition is performed using the signal addition count shown in FIG. In other words, for the first time, only the entire excitation is performed with a 90' pulse, and at that time, the FID (No. 3 is set as positive), and for the second time, a 180° pulse is given for X-axis selection, and 180° is given for each of Y-axis selection and Z-axis selection. A 90° excitation pulse is applied without applying a ° pulse (this is applied any time), and the FID signal obtained at this time is set as negative.Hereafter, 8 FrD signals obtained in the same sequence of 8 times By adding with the given polarity, the signals in the non-selected parts cancel each other out and are not output, and the X-axis.

The FID signals of only the Y-axis and the portions (selected portions) intersecting each selected plane of the Y-axis are added and output. In addition, in this method, the spin rotation angle (flip angle) due to each 180° pulse of x, y, and z is all 180°.
It has been theoretically shown that at any angle α, β, or γ, all signals from outside the selected portion are canceled.

■ The pulse sequence in 5PAR5 is the 13th
As shown in the figure, this can be said to be an improved version of the VSE shown in (3). That is, the non-selective 90° pulse and the selected 45°+45° pulse in the VSH slice selection of (2) are separated, and the non-selective 180' pulse is inserted between them. By making these three pulse intervals equal, a spin echo occurs at the center of the selected 90° (45° + 45”) pulse, correcting the phase variation due to Bo nonuniformity and chemical shift. selection 180
The gradient magnetic field pulse G1 immediately before the pulse is for phase compensation of the slice selection gradient magnetic field pulse axS that is applied together with the selection 90° pulse.

■ To further elaborate on 5PAR5, non-selection 90
゛The entire subject is excited by the X pulse, and each excited spin generates a phase difference over time in accordance with the difference in frequency. When a 180°y pulse is given,
The spins are exchanged within the plane of rotation around the phase center of each spin, and the one that was rotating fastest until then becomes the position of the slowest spin. Therefore, when the time interval between these 90° When applying a slice selection 90°
The phase is separated by t.

If you do the same for the Y-axis and the Y-axis, the spin of only the part that intersects the three selected planes will be the Y-axis or -
This is in the Z-axis direction, and when a 90° excitation pulse is applied, only this selected portion is excited.

There are two methods described in JP-A-59-107246 (1), and the pulse sequence of the first method is shown in FIG. This method applies a high-frequency pulse that saturates a certain selected area outside the chamfer under a first gradient magnetic field G8, in other words, the spin is apparently invisible, and then a second gradient magnetic field Gy Below, a 90° selection pulse is applied to excite a plane perpendicular to a specific point on the Y axis in the unsaturated plane (selected plane), and then under a third gradient magnetic field G8, a 180° selection pulse is applied. A pulse is applied, and only the spins on the plane orthogonal to a specific point on the X-axis are replaced with leading spins and lagging spins, and spin echo signals at the intersections of the three selected planes are extracted.

The pulse sequence in the second method of (2) is shown in FIG. 15. In this case, first, a 90° selection pulse is used to
Excite spins within the selected slice plane. Next 18
The 0° selection pulse is repeated to spoil spins other than the selected portion in other selected slices, and then spin echo signals from the intersections (selected portions) of the three selected slice planes are observed.

[Problem that the invention seeks to solve]

These conventional techniques have the following problems.

In the VSE method described in (2), it is necessary to apply a non-selective 90° pulse under a gradient magnetic field.To do this, the pulse width must be made sufficiently short to widen the frequency range to be generated, and it is necessary to apply a non-selective 90° pulse. It must be excited in correspondence with each magnitude of the gradient magnetic field, and in order to transmit this narrow 90° pulse, the transmitter output needs to be large, and the peak value also affects the subject, especially the human body. This results in the application of a high high frequency power, which is undesirable, and furthermore, longitudinal relaxation (T1) may not be negligible between the application of the non-selective 90° pulse and the application of the excitation 90° pulse.

In the l5zS method of (2), the FID signal actually observed each time has many signals superimposed on it from areas other than the parts that should be selected, resulting in a signal with a high level as a whole.As a precondition for digitally processing this, Although an expensive AD converter with a large dynamic range must be used as an AD converter, there is a limit to this, and the S/N ratio of the resulting signal from the selected area deteriorates.

In the 5PAR3 method (2), the number of high-frequency pulses used is large compared to other conventional methods, so the time from the first 90° As a result, the FID signal deteriorates. Also, a non-selection pulse 90”x.

Since a large number of 180° y angles are used, there is a drawback that the high frequency power applied to the subject also increases.

The first method described in Japanese Patent Application Laid-Open No. 59407246 uses a powerful long saturated high-frequency pulse with a frequency component selected to irradiate the entire area other than the selected plane, so that it is absorbed by the subject. The high frequency power is also large (which is undesirable).In addition, in the second method, in order to observe the spin echo signal, the time from the first 90° selective pulse in Fig. 13 to the reception of the desired spin echo signal is There is a limit to this, and it cannot be shortened.Furthermore, since there is no spooling in two directions, there is a risk that signals from areas other than the selected portion may not be removed.A 90° pulse component occurs at the rising and falling edges of a 180° pulse. , it is difficult to get the correct signal.

[Means for solving problems]

According to the first aspect of the invention, a static magnetic field is applied to an object to excite specific nuclear magnetic spins in a selected first cross section of the object, and then first ．． Second. The magnetic field strength changes gradually along the third axis,
1. The direction of the magnetic field is parallel to the static magnetic field. Second. A third gradient magnetic field is applied to the subject. Thereafter, a radio frequency 180° is applied to rotate the specific nuclear magnetic spins by 180° in a selected second cross section that intersects with the selected first cross section.
A pulse is applied to the object, and then the first . Second. The fourth axis has a magnetic field strength varying along the third axis, and the direction of the magnetic field is parallel to the static magnetic field. Fifth. applying a sixth gradient magnetic field and then applying a selected third magnetic field that intersects the selected first cross section and the second cross section;
A high frequency 180° pulse is applied to the subject to rotate the specific nuclear magnetic spins by 180° in the cross section of the first, second, . The magnetic field strength varies along the third axis, and the direction of the magnetic field is parallel to the static magnetic field in the 7th and 8th directions. A ninth gradient magnetic field is applied. After that, the above-mentioned 1. Second. A nuclear magnetic resonance signal (free induction decay signal) from a selected portion intersected by the third cross section is received, converted into a digital signal and processed to obtain nuclear magnetic resonance spectroscopy information of the selected portion.

Since the first aspect of the invention is configured as described above, the first aspect of the invention is as follows.
．． Second. When the third (IJI) oblique magnetic field is applied, the phases of the motion of the spins excited by these magnetic fields become disjointed, resulting in so-called skimming.The first 180°
Upon application of a pulse, in a second cross-section intersecting the first cross-section, the disintegrated spins have their rotational direction reversed, thus causing the fourth. Fifth. The integral value of the energy of the sixth gradient magnetic field is determined by the integral value of the energy of the sixth gradient magnetic field. Second. When the magnetic field matches that of the third gradient magnetic field, the spin of the rod-shaped portion where the first cross section and the second cross section intersect returns to the excited state. These are also the fourth. Fifth. The remaining part of the sixth gradient magnetic field is skimmed, and the application of the next 2180° pulse causes the first... Second. The same switching action as described above is performed for the spins at the intersection of the third cross section, that is, the intersection (selected portion) of the rod-shaped portion and the third cross section. After that, the 7th. 8th.
A ninth gradient magnetic field is applied, and the integral value of this energy is
Said No. 4. Fifth. The energy is made equal to the energy at the time of boiling by the remaining part of the sixth gradient magnetic field, so that the spins of only the selected part are returned to the excited state, and an FID signal is emitted from the returned excited spins, and the nuclear magnetic resonance spectrometer Scope information can be obtained.

According to the second aspect of the invention, a static magnetic field is applied to an object to excite specific nuclear magnetic spins in a selected first cross section of the object, and then nuclear magnetic spins that are not in the same plane and intersect with each other are excited. 1.
Second. The magnetic field strength varies gradually along the third axis, and the direction of the magnetic field is parallel to the static magnetic field. Second. Third
A gradient magnetic field is applied to the subject. Thereafter, a high frequency wave 180 is applied to rotate the specific nuclear magnetic spin by 180° in a selected second cross section that intersects with the selected first cross section.
° pulses are applied to the subject, and then the first .degree. Second. A fourth . Fifth. Applying a sixth gradient magnetic field, the specific nuclear magnetic spins in the selected third cross section intersecting the selected first cross section and the second cross section are 180 degrees.
A high-frequency 180° rotating pulse is applied to the subject, and then the first, second, and second . The magnetic field strength varies along the third axis, and the direction of the magnetic field is parallel to the static magnetic field in the 7th and 8th directions. A ninth gradient magnetic field is applied. Thereafter, a tenth gradient magnetic field whose magnetic field strength is gradually changed along the fourth axis and whose magnetic field direction is parallel to the static magnetic field is applied as a reading gradient magnetic field to receive a nuclear magnetic resonance signal (free induction signal), and a nuclear magnetic resonance signal (free induction signal) is received. An eleventh gradient magnetic field whose magnetic field strength gradually changes along a fifth axis intersecting the fourth axis and whose magnetic field direction is parallel to the static magnetic field is applied from after the excitation of the nuclear magnetic spins to before the reading gradient magnetic field is applied. is applied for a short period of time, and the gradient of the magnetic field strength of the eleventh gradient magnetic field is changed to repeat the process from excitation of the nuclear magnetic spins in the first cross section to reception of the nuclear magnetic resonance signal, and each received magnetic resonance signal is The detected nuclear magnetic resonance signal is converted into a digital signal and processed.

Since the second aspect of the present invention is configured in this manner, it operates in substantially the same manner as the first case of the present invention, and furthermore, since the first O1 and eleventh gradient magnetic fields are used, the 4. In-plane nuclear magnetic resonance image information constituted by the fifth axis is obtained.

〔Example〕

First, a typical embodiment for obtaining nuclear magnetic resonance spectroscopy information, which is the first aspect of the present invention, will be described. 1st
In period 1 in the figure, the magnetic field direction is the main magnetic field (static magnetic field) direction (Z-axis direction), and a high-frequency selection 90° pulse is applied to the subject 1 while applying a gradient magnetic field Gy whose intensity has a gradient with respect to the Y-axis.
1.

The subject 11 is arranged in a rectangular coordinate system X, Y shown in FIG.

Suppose that it is placed in Z. Therefore, the nuclear magnetic spins in the X-2 slice plane (cross-section) 12 in FIG. 2 are excited by G in the opposite direction in period 2 in FIG. 1 at the 0th order.

Go (referred to as a rephasing magnetic field) is applied to align the phase shift in the Y direction of the nuclear magnetic spins excited in the X-Z slice plane.

In this invention, in period 4, large G −, G v, G
Apply z. G is a gradient magnetic field (referred to as an X-axis gradient magnetic field) in which the magnetic field strength gradually changes along the X-axis and the magnetic field direction is parallel to the Y-axis, and Gv is a gradient magnetic field in which the magnetic field strength gradually changes along the Y-axis, These large cx, cv, and cz, which are gradient magnetic fields whose magnetic field direction is parallel to the Y axis (denoted as 'l gradient magnetic fields), are 01
It is written as 111 0 Yl+ G z*. This GXll
The phase φ of the nuclear magnetic spin excited by the application of G Vl+ G z* is as follows according to each position (x, y, z).

Period 4 Period 4 Period 4 Here, gx=gy2gs is the gradient magnetic field G
+G□ and G25 intensities. Therefore, the phases of the nuclear magnetic spins excited in period 4 become scattered.
This is called boiling.

Next, in period 6, by applying a high-frequency selective 180° pulse while applying a gradient magnetic field G, the relationship between the selected plane perpendicular to the X axis in the X-Z slice plane 12 as shown in FIG. The nuclear magnetic spins of the crossed rod-shaped region 13 are reversed in the direction of the high-frequency 180° pulse in the spin rotation coordinate system within a plane perpendicular to the direction of the static magnetic field (Z-axis direction),
In other words, the rotational direction of the spins is reversed, and within the slice plane selected by the high-frequency 180' pulse,
The nuclear magnetic spin in the specified portion excluding the rod-shaped region 13 in FIG. 3 is in the opposite direction to the direction of the static magnetic field.

Next, by applying the same gradient magnetic fields G, , G□, G23 as in period 4 in period 8 in FIG. The phase becomes even larger by the same amount as the phase φ given by equation (1), and the proportion of boiling becomes larger than in period 4. However, since the direction of the nuclear magnetic spin in the rod-shaped region 13 is reversed in the period 6, the phase of the nuclear magnetic spin in the rod-shaped region 13 changes in opposite directions in the period 8 by the phase given by equation (1). Therefore, the phase applied in period 4 is canceled out in period 8, and the state returns to the returned excited state.

Furthermore, if the high-frequency pulse is non-ideal, such as being spatially non-uniform, for example, the high-frequency 180° pulse is incomplete and the nuclear magnetic spins in the portion other than the rod-shaped region 13 may move in the main magnetic field direction (Z-axis Even if the nuclear magnetic spin has a component perpendicular to the direction), the phase of the nuclear magnetic spin in the portion other than the rod-shaped region 13 is scattered in the period 8, that is, it is spoiled. That is,
It is only in the bar-shaped region 13 that the boiling caused by periods 4 and 8 is canceled out.

Next, in period 10, a gradient magnetic field G z @,
Gva+GZs is applied to skim the phase of the nuclear magnetic spin, and a gradient magnetic field G is applied in period I2.

By applying the second high-frequency selective 180° pulse while applying the The nuclear magnetic spins are reversed with respect to the orientation of the high-frequency 180° pulse in the rotational coordinate system of the spins in a plane perpendicular to the direction of the static magnetic field, and within the slice plane selected by the second selection 180° pulse. The specific nuclear magnetic spins in the portion excluding the rod-shaped region 14 in FIG. 3 are in the opposite direction to the direction of the static magnetic field.

Next, in period 14, the same gradient magnetic field G as in period 10,
,,G□*Gzs, the phase of the nuclear magnetic spin in the portion other than the rod-shaped region 14 is further increased by the same amount as the phase φ given by the il1 equation. However, since the direction of the nuclear magnetic spin is reversed in the period 12, the phases of the late magnetic spins in the rod-shaped region 14 change in opposite directions in the period 14 by the phase given by equation (1). Therefore, the phase applied in period 10 is canceled out in period 14. For regions 14a and 14b in the bar-shaped region 14 that do not intersect with the bar-shaped region 13, even if the skimming in periods 10 and 14 cancels out, the effects of skimming in periods 4 and 8 still exist.

Therefore, in the period 14, the sboiling is completely canceled out only in the region (selected region) 15 where the bar-shaped regions 13 and 14 intersect with each other.

Next, in period I5, in the absence of a gradient magnetic field, the NMR signal (FID
signal) is emitted, and this NMI? By observing the signals and performing Fourier transformation, a nuclear magnetic resonance spectrum of the selected region 15 in the subject is obtained.

Further, in order to obtain the signal from the selection area 15 in FIG. 3, the sequence shown in FIG. 4 may be used. High frequency selection 90” pulse in period 1 and high frequency selection I in period 6 in Figure 4
The period between the high-frequency selection 180° pulse in period 6 and the high-frequency selection 180° pulse in period 12 is set to τ1 + τ2. Selected 180° pulse to period τ
2, a spin echo signal is emitted by the nuclear magnetic spins in the selected region 15 in FIG. 3, and by observing this signal and performing Fourier transformation, a partially selected nuclear magnetic resonance spectrum of the object is obtained.

Periods 2 to 5, periods 7 to 11, and periods 13 to 1 in Figure 1
14 may be executed all at once as in periods 2.4.6 as shown in FIG. 5. However, the time integral of each gradient magnetic field is 5
Equal to period 2 in the figure, and periods 7 to 11 in figure 1
and period 4 in Figure 5 are equal, and period 1 in Figure 1 is equal.
Each value is selected so that period 3 to 14 and period 6 in FIG. 5 are equal.

In addition, the gradient magnetic field strength in periods 4 and 8 in Fig. 1 and period 1
The gradient magnetic field strengths of 0 and 14 do not need to be the same value, and can be arbitrarily set as necessary.

In the second embodiment of the present invention, based on the nuclear magnetic resonance signals from the selected area 15 in FIG. 3, NMR signals are collected from this area 15 so as to convert NMR information into an image. For example, as shown in FIG. 6, periods 1 to 13 perform almost the same processing as periods 1 to 14 in FIG.
4, a phase encoding gradient magnetic field (warb gradient magnetic field) and a defending gradient magnetic field G are applied, and a reading gradient magnetic field is applied in period 15. Period 14 includes periods 2 to 4, periods 6 to 10, and period 12. It may be set to any period from 1 to 13, and the period between each high-frequency pulse and periods 12 to 14 may be executed all at once.

A spin echo image of the selected region 15 in FIG. 3 can also be obtained by replacing period 15 in FIG. 4 with period 14.15 in FIG. 6. Furthermore, by applying a non-selective 180° RF pulse for a period before period 1 in FIG. 6, a longitudinal relaxation-enhanced image of the selected region 15 in FIG. 3 can be obtained.

In order to obtain a chemical shift image of the selected region 15 in FIG. 3, the sequence shown in FIG. 7 can be carried out. In FIG. 7, the method shown in Japanese Patent Application No. 194621/1988 is applied to period 14 and period I5 in FIG. Apply c.

FIG. 8 is an example in which the encoding and reading portions of a conventionally known three-dimensional imaging sequence are added to periods 14 and 15, and as a result, the nuclear magnetic resonance three-dimensional image of the selected area 15 in FIG. is obtained. By adding the encoding and reading part of the three-dimensional chemical shift imaging pulse sequence described in JP-A-59-108946 to period 14.15 in FIG. 9, three-dimensional chemical shift imaging of the selected region 15 in FIG. is obtained.

〔Effect of the invention〕

As described above, according to the first aspect of the present invention, a nuclear magnetic resonance spectrum of an arbitrary part of a subject can be obtained, and according to the second aspect of the present invention, a nuclear magnetic resonance image can be obtained.

In this case, since a high frequency selective pulse is used, it can be carried out with less high frequency power compared to a method using a high frequency non-selective pulse, reducing the burden on the transmitter and the influence on the subject. I can do it.

Furthermore, since the signal is extracted only from the selected portion, the dynamic range problem of the AD converter can be effectively reduced compared to the conventional 15iS method.

Furthermore, since the time from excitation to obtaining a signal can be shortened, the influence of relaxation of nuclear magnetic spin can be effectively reduced.

Because the boiling is performed using each gradient magnetic field,
Nuclear magnetic spins other than those to be selected can be effectively skimmed.

Furthermore, in order to obtain the desired S/N, averaging is often performed, but when performing partial selection spectroscopy,
By carrying out the method shown in FIG. 8, a nuclear magnetic resonance spectrum can be obtained more effectively. For example, n
When averaging three times is required, it is better to perform a sequence to obtain nuclear magnetic resonance spectra of 03 voxels (for example, when the resolution in the vertical, horizontal, and depth directions is n).
Spatial information increases, i.e., 1 obtained by the method shown in Figure 8.
Since the S/N of the nuclear magnetic resonance spectrum of one voxel is equal to the S/N obtained by repeating the partial selection sequence of that one voxel 02 times, it is better to use the method shown in Fig. Obtaining nuclear magnetic resonance spectra for , which increases spatial information.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing a sequence example of the first partially selective nuclear magnetic resonance image deriving method of the present invention, and FIG.
Figure 3 is a diagram showing the relationship with the Z slice plane, Figure 3 is a diagram showing an example of the relationship between the X-Z slice plane and the partially selected region, and Figure 4 is partially selective nuclear magnetic resonance information derived using the spin echo method. FIG. 5 is a diagram showing an example of a sequence in which the time axis is shortened in the method of the present invention, and FIG. 6 is a diagram showing an example of the second partially selective nuclear magnetic resonance image derivation method of the present invention. Figure 7 is a diagram showing an example of a pulse sequence that is used in the method of obtaining a partially selective nuclear magnetic resonance chemical shift image, and Figure 8 is a diagram showing a partially selective nuclear magnetic resonance 3 sequence.
FIG. 9 is a diagram showing an example of a sequence applied to a method of obtaining a 3-dimensional image. FIG. 9 is a diagram showing an example of a sequence applied to a method of obtaining a partially selective nuclear magnetic resonance three-dimensional chemical shift image. FIG. Diagram showing the sequence of the conventional VSE method, 1st
Figure 1 shows the sequence of the conventional 1515 method.
Figure 2 shows the execution order of the sequence of the conventional rsTs method, Figure 13 shows the sequence of the conventional 5PAR3 method, and Figure 14 shows the first method described in JP-A-59-107246. FIG. 15 is a diagram showing the sequence of the second method described in JP-A-59-107246.

Claims (2)

[Claims] (1) Place the subject in a static magnetic field, excite specific nuclear magnetic spins in the selected first cross section of the subject, and then make sure that the direction of the magnetic field is the same as the static magnetic field, and a first gradient magnetic field whose intensity gradually changes along a first axis; a second gradient magnetic field whose magnetic field intensity gradually changes along a second axis different from the first axis; a third gradient magnetic field in which the magnetic field strength gradually changes along a third axis in a plane different from the first and second axes, is applied to the subject, and then the selection of the subject is performed. The specific nuclear magnetic spin in the selected second cross section that intersects the first cross section is 1
A first high-frequency 180° pulse rotating by 80° is applied, and then a fourth gradient magnetic field is applied in which the direction of the magnetic field is the same as the static magnetic field and the magnetic field strength is gradually changed along the first axis. , a fifth gradient magnetic field whose magnetic field strength gradually changes along a second axis different from the first axis, and a third gradient magnetic field whose magnetic field strength is in a plane different from the first and second axes. A sixth gradient magnetic field that is gradually varying along the axis is applied to the subject, and then a sixth gradient magnetic field that intersects the selected first cross-section and the selected second cross-section of the subject is applied to the subject. A second high-frequency 180° pulse is applied to rotate the specific nuclear magnetic spins in the cross section of 3 by 180°, and then the direction of the magnetic field is in the same direction as the static magnetic field, and the magnetic field strength is along the first axis. a seventh gradient magnetic field whose magnetic field strength gradually changes along a second axis different from the first axis;
and a ninth gradient magnetic field whose magnetic field strength is gradually changing along a third axis that is in a plane different from the first and second axes, are applied to the subject, and then, A partially selective nuclear magnetic resonance information deriving method, which receives a free induction signal generated from the subject and derives desired information at an intersection of the first, second, and third cross sections. (2) Place the subject in a static magnetic field, excite specific nuclear magnetic spins in the selected first cross section of the subject, and then make sure that the direction of the magnetic field is the same as the static magnetic field, and a first gradient magnetic field whose intensity gradually changes along a first axis; a second gradient magnetic field whose magnetic field intensity gradually changes along a second axis different from the first axis; a third gradient magnetic field in which the magnetic field strength gradually changes along a third axis in a plane different from the first and second axes, is applied to the subject, and then the selection of the subject is performed. The specific nuclear magnetic spin in the selected second cross section that intersects the first cross section is 1
A first high-frequency 180° pulse rotating by 80° is applied, and then a fourth gradient magnetic field is applied in which the direction of the magnetic field is the same as the static magnetic field and the magnetic field strength is gradually changed along the first axis. , a fifth gradient magnetic field whose magnetic field strength gradually changes along a second axis different from the first axis, and a third gradient magnetic field whose magnetic field strength is in a plane different from the first and second axes. A sixth gradient magnetic field that is gradually varying along the axis is applied to the subject, and then a sixth gradient magnetic field that intersects the selected first cross-section and the selected second cross-section of the subject is applied to the subject. A second high-frequency 180° pulse is applied to rotate the specific nuclear magnetic spins in the cross section of 3 by 180°, and then the direction of the magnetic field is in the same direction as the static magnetic field, and the magnetic field strength is along the first axis. a seventh gradient magnetic field whose magnetic field strength gradually changes along a second axis different from the first axis;
and a ninth gradient magnetic field whose magnetic field strength is gradually changing along a third axis that is in a plane different from the first and second axes, are applied to the subject, and then a tenth gradient magnetic field is applied to the subject. while applying a gradient magnetic field as a reading magnetic field, receiving a free induction signal generated from the object, and obtaining an eleventh gradient from after the excitation of the nuclear magnetic spins in the first cross section to before applying the gradient magnetic field. A magnetic field is applied, and the process from the nuclear magnetic spin excitation of the first cross section to the reception of the free induction signal is repeated while changing the intensity, and the received free induction signal is used to generate the first, second, and A partial selective nuclear magnetic resonance information derivation method for deriving desired information from the intersection of the three planes.