SU1467476A1 - Method of tomography based on nmr - Google Patents

Abstract

The invention relates to nuclear magnetic resonance tomography and can be used to visualize the spatial structure of the objects under study. The purpose of the invention is to reduce the measurement time. The object under study with a characteristic size L and a cniiH lattice relaxation time T is placed in a constant magnetic field H with a relative inhomogeneity of the LF, is exposed to it with a 90 selective RF pulse in the presence of a Z gradient of the magnetic field, then iDiTHoro field, register the NMR signal in the presence of X-degree magnetic field and repeat scanning cycles with different cells and Y-rpa magnetic field, turn on simultaneously with the Y-gradient magnetic field through the interval C -10%, where N is the size of the NMR matrix of the tomogram, turn off the Y-gradient of the magnetic field and act on the object with a sequence of 180 radio-frequency impulses (180 ° -, - .-)) n, where n 1 - N - the cycle number, scanned, G, b, register the NMR signal in the current interval, vmesh-1 €, and during the interval about, they affect the Y-gradient of the map field, the X-gradient of the magnetic field being constant and equal to G AHNHo / L and the magnitude of the Y-gradient magnetic field is determined from the conditions: j Gy (t) dt & amp & HCHo / L. 5 il. about oh

Description

The invention relates to the field of nuclear magnetic resonance tomography and can be used to visualize the spatial structure of the objects under study.

The aim of the invention is to reduce the measurement time.

FIG. 1 shows the timing diagram of the delivery of radio pulses; in fig. 2-4 - diagrams of the supply of magnetic field gradients along the axis m Z, X and Y

respectively; in fig. 5 shows diagrams of nuclear magnetic resonance (NMR) signals.

The method of computational tomography based on nuclear magnetic resonance is carried out as follows.

When exposed to a test object, placed in a permanent magnetic field, a 90% selective radio frequency pulse (Fig; 1) in the presence of a Z-gradient magnetic field (Fig. 2) excites only the spins of the nuclei located in the perpevicular plane The p1 axis is Z. The width of the selected layer depends on the spectrum of the 90 ° selective RF pulse and the magnitude of the Z-gradient of the magnetic field. The direction of the layer selection is chosen purely arbitrarily, in the same way you can select a layer in any direction, including the X and Y-magnetic field gradients (Figs. 3 and 4) or any 1-line combination of magnetic field gradients.

All further reasoning / day relates to a two-dimensional layer lying in the XYo plane. In order to obtain the data needed for fjfln to obtain the HH5j NMR waveform, enter the coding of the resonant conditions along the X axis - using the X gradient of the magnetic field - G | ((t) and Y axis - using the Y-rpa magnetic field Gu (t). For example, let the object under study be placed in a constant magnetic field HQ, then the resonance conditions are written as follows:

where CO2 is the resonant frequency of the NMR signal;

V is a gyromagnetic ratio. In addition, a magnetic field gradient is imposed on the object under study, for example, Gj (t)., Then

And id + g, (t) x,

1 de ss is the coordinate of the point of the object under study.

Therefore, the resonance conditions are now defined as:

And + G; (t) X - QO + J G; (t) .x.

Thus, the resonant frequency depends on the x coordinate, the trace-signal spectrum of the NMR signal is now.

contains information that has been removed with the coordinate of any point X of the object under study. Under real conditions, a constant magnetic field H is ideal and has heterogeneity, denoted by its & H. Then, in order to separate the spectrum of the NMR signal of the K points, it is necessary that the condition

L G, (t) & H N,

where L is the characteristic size of the object under study; N is the size of the NMR tomogram.

Hence the condition determining the magnitude of the gradient:

Gv (t) i

L

G point of view for obtaining the maximum signal-to-noise ratio requires that the gradient of the magnetic field Gj (t) be as small as possible, since the width of the spectrum of the NMR signal increases and the signal-to-noise ratio decreases as the gradient increases.

Thus, the optimal magnitude of the magnetic field gradient is determined from the ratio

G, (t)

L

All reasoning is true for Gu (t). However, the proposed method of encoding resonant conditions is valid only for the one-dimensional case. In this way, it is not possible to encode both in X and Y, since with simultaneous switching of two gradients, they are cy and Fcc and one-dimensional encoding is obtained again, but only along the total axis.

In the proposed method, the NMR signal coding, necessary for obtaining the specified matrix, is performed using a spin echo, which is generated by a sequence of 180-rf RF pulses (Fig. 1):

(180 - e, -) p,

where f-, and - time intervals; n is the scan cycle number.

The magnitude of the time intervals is 2, and is determined by the ratio

/

J bin Til C ibiests from those yc.noBiii i, so that the amplitude of the NMR signal is reduced by no more than / e in the total registration time, therefore,

2 GM T can be determined

 -

. 2N

The duration can be selected and less, however, this entails an increase in the amplitude of the magnetic field gradients, which causes an increase in the reception bandwidth of the NMR signal and, therefore, increases the noise.

The pause between the TZO-s radio frequency pulses is equal to 2, specially divided into two interpals. During the first interval t, the NMR signal is recorded, and during

about

the time of the second is undergoing the phasing process. It should be especially noted that the method uses the effect of accumulating the coding phase along the Y-axis due to the cy-world of the selective effect in the 180-sequence interval. radio frequency pulses. This reduces the Y-gradient of the magnetic field by more than two orders of magnitude. On the other hand, the gradients of the magnetic field G, GK and GU in this case are constant and, therefore, to implement it, it is not necessary to have a complex control system that changes the amplitude, for example, the C-gradient in accordance with a given function. Control in the proposed method is carried out in a key mode. As already shown, in order to resolve K points, it is necessary to have the graph “LRN

Dient magnetic field equal to -.

Lj

Therefore, the increment of the coding phase during each scan cycle should be

LNS2N „

In both types, uHtH can be written.

J

C. (t) dt

if Gu (t) -const, then

. - ,, -, at

DAYS

t

s

five

0

five

0

five

0

About I K.:HMltM H about PMI. W T y .1 I-l- and I-1 pl. Reader NfnrnHTiioro scrap from;) (1Y makes About S1Gku on coordinatepayts in the NMR frame. It is believed that the accuracy of the coordinates of the coordinate in the NMR tomogram it should be ok; and it should be within the limit of 1%, which is due to); about both the tyPap} ibLMH factors, and the accuracy of the calculation. Already sufficiently complete elimination of artifacts caused by a deviation of a certain amplitude of gradients from a given value, it is necessary that this deviation be no more than 0.1%.

I

PRI mere Explore the object

over time, the spin-spin realization 1 and the characteristic size L are placed in a constant MarniiTiioe field with an intensity of 0.14 T and a relative inhomogeneity of 5x10. The duration of the 90th selective radio frequency pulse is 10 T / m. The size of the NMR tomograf matrix; The duration of the 180 ° RF pulse is 3x10 s.

The use of the method will allow to reduce the scanning time of P1 by 300 times in comparison with P1u with a known method. The magnitude of the Y-gradient of the magnetic field is 128 times as large as the magnitude of the magnetic field, and, consequently, the power of the magnetic field of the magnetic field is 10,000 times less. In addition, .X, Y, r-gradients of the magnetic field have a strictly defined amplitude, which does not change in magnitude; this greatly simplifies the control system and gradient-eHTaNm magnetic field. The total scan time is 4x10 sec.

Claims (1)

Invention Formula The method of nuclear magnetic resonance (NMR) based tomography is stored in that a test object with a characteristic size L and a spin-lattice relaxation time T is placed in a constant magnetic field. field with relative heterogeneity effect of the DN, affect it with a 90 ° selective RF pulse in the presence of a Z gradient of the magnetic field, then apply a Y gradient of the magnetic field, record the signal in the presence of the X gradient of the magnetic field, and repeat exposure cycles at different 146 values of the Y-gradient of the magnetic field, according to the data obtained, an NMR tomogram is computed, characterized in that, in order to reduce the measurement time, the X-gradient of the magnetic field is simultaneously included in the Y-gradient of the magnetic field through an interval of time - ± 10% where N NMR tomography matrix size, turn off the Y-gradient of the magnetic field and additionally affect the object with a sequence of radio frequency impulses (180 °, .-) p, 6. 8 where n 1 - N is the number of the cigspar; , Oj 2, record the NMR signal during the time interval p1, and during the interval about, the Y-gradient of the magnetic field is affected, the X-gradient of the magnetic field being constant and equal to LN - K Nd, and the magnitude of the Y-gradient magnetic field is determined from the conditions 15 jG (t) dt , 180  G; tz K gradiyatp  rni; ---, - --- i-Ji-J | Sch || D-iiU-LLui -LL-ijJUMHwi 180