SU1368752A1 - Method of nmr-tomography - Google Patents

Abstract

The invention relates to computational tomography methods based on nuclear magnetic resonance, intended for introscopic studies in medical diagnostics and non-destructive testing. The purpose of the invention is to increase the accuracy of research by increasing the signal-to-noise ratio. The essence of the method lies in the fact that the object of study twice in the allotted time of study is exposed to a radio pulse that excites a damped free induction signal and an echo signal. In this case, the phases of the first half of the harmonic oscillations are set to O or 180, and the second half to 90 or 270, and the resulting double exposure signals are averaged. 8 il. (L

Description

with about

00

ate IN

The invention relates to methods for reconstructive computed tomography based on the phenomenon of nuclear magnetic resonance (NMR) and is intended for quantitative introscopic studies of various media, the results of which are used for the purpose of medical diagnostics and non-destructive testing.

The purpose of the invention is to increase the accuracy of the study by increasing the signal-to-noise ratio.

Figure 1 shows the block diagram of the device for implementing the method in figure 2 and 3 - vector diagrams of radio pulses; Fig. 4 shows a plot of stresses allowing the method to be carried out; Figures 5-8 show the signals at the output of the quadrature detector U (t) and V (t) and after the Fourier transform with phase corrections R (to) and I (a).

The essence of the invention is to use an excitation radio pulse, leading to the appearance of quadrature signals of a free induction decay (FID) and echoes, and in the allotted time of the study to scan the object twice.

The device contains an electromagnetic coil unit 1 comprising a main magnet coil, a radio frequency katuatka and coils creating a magnetic field using X, Y and Z modes, and intended for placing the object under study in its inner space, the coil control unit 2 of the preunit 10 forming echo excitation intervals; Moreover, the output of the interval shaping unit 10

excitation of echo signals is connected to a programmable pulse generator 11, one output of which is connected to the clock input of the analog-digital converter 6 and the other to the input of the radio pulse shaping unit 12, the output of which is connected to the input of the transmitter 13. On the output interfaces of the computer 8 is connected to the block 10 generating echo excitation intervals, with a halftone display 9 for visualizing the received images, and a magnetic field gradient formation unit 14, the outputs X, Y and Z of which

connected to a multichannel digital-to-analog converter 15, the analog outputs of which are connected to a pre-amplification unit 1 6, the outputs X, Y and Z of which are connected to

unit 3 controls the magnetic field gradients.

The method is carried out as follows.

An image acquisition method is selected, for example, by projections. The object of investigation is placed in a block of magnetic coils and fixed there. Select the ICCJ plane of the section cross section. Forward from the computer 8 to the magnetic field formation unit 14 the coordinates of the inclination of the plane of the section under investigation. In block 14, gradient formations generate gradient values along the axes X, Y, and Z, with

assigned to maintain a voltage of 40 corresponding to the orientation of the selected field of a given magnitude, block 3 of the plane and different directions of controlling the gradients of the magnetic field along the axes X, Y and Z, passive

switch 4, which is

projection. The gradient values corresponding to the first projection angle are fed to the digital-analog assembly of the diodes, the resistance of which is 45 converter 15, where the transition depends on the direct analog form supplied to it, is amplified by the block

16 pre-amplify and enter gradient control block 3.

magnetic field from which output

Nor, and designed to protect the input circuits of the quadrature detector 5 from the powerful pulses of the transmitter, the switch is connected between the radio frequency gg and is fed to the gradient coils by the magnetic coil of block 1, by the quadrature field of the block 1 of the coil electromagnetism.

At the same time, in block 10, with a sensor 5 and a transmitter, an analog-to-digital converter 6 connected to the outputs of the quadrature detector 5, a buffer memory 7 for storing digitized echo signals and connected between the analog-to-digital converter 6, the computer unit 8, the semitone display 9 and

55

On the excitation intervals, an echo signal is given to a command from the computer 8 to form a sequence with spread (180) echo excitation pulses. The generated sequence is loaded into

magnetic field from which output

Received on the coils of the gradients of the magnetic field of the block 1 electromagnet

On the excitation intervals, an echo signal is given to a command from the computer 8 to form a sequence with spread (180) echo excitation pulses. The generated sequence is loaded into

3

a programmable pulse generator 11, from the output of which, on command from a computer 8, enters a radio pulse shaping unit 12, where radio pulses are generated (90 °) with the required set of frequencies and phases determined by the Hadamard coding matrix, as well as wideband radio pulses (180) the transmitter unit 13, where it is amplified, and through the passive switch 4 is fed to the radio frequency coil of the electromagnetite coil unit 1 (Figures 2 and 3),

Simultaneously, the programmable pulse generator 11 starts producing clock pulses arriving at the clock input of the analog-digital converter 6. Upon the arrival of these pulses, the analog-digital converter 6 digitizes the signals detected by the quadrature detector 5. The digitized signals are stored in the buffer memory block 7 and accumulate. Plots 17-21 gradients and observed signals are shown in FIG. 4.

Further, these actions are repeated again using the same gradient values, but using the following line from the Hadamard coding matrix. So they do until all the rows of the matrix are sorted out. As a result, a set of complex data (matrix) is obtained. A complex Fourier transform of the obtained values of the complex data is carried out for the corresponding layers and a preliminary complex value of the projections is obtained for the first projection direction in two layers. To obtain the values of the projections, corresponding to other layers, a Fourier transform and phase correction are produced. Rejection of signals corresponding to other orthogonal pairs of layers.

Next, information about the next projection is collected. From the computer 8, the command to establish the projection is sent to the magnetic field gradient formation unit, and the signal acquisition-processing operation is repeated. -Such actions Produce up to as long as they do not go over all the values of the gradients and receive a complete set of 5X data for each section under study.

68752 i

A rescan is then performed to completely repeat the original. The results of the two scans were averaged. The resulting data set is processed by the reconstruction algorithm and visualized on a halftone display 9.

Thus, in the proposed method, the study is conducted twice during the allotted time, and the results are averaged, which leads to an increase in the signal-to-noise ratio by V2 times, without complicating the device to implement the method and complicate the processing algorithm.

Claims (1)

Invention Formula 20 An NMR tomography method consisting in placing an object of study in a constant magnetic field, applying magnetic field gradients, exciting free induction signals with a radio pulse consisting of a set of modulated harmonic oscillations, frequencies that determine the position of the layers under study, and the phases are set to O or 180 ° in accordance with the Hadamard coding matrix, excitation of echoes by a wideband radio pulse, detection of echoes by a quadrature detector, digitization and accumulation of digitization signals with repeated repetition of the above actions to obtain a complete set of data, which is processed by solving a system of linear equations separating the signals from different layers of the object, processing the separated signals by a reconstruction algorithm and visualizing the results of the reconstruction, in order to increase the accuracy of the investigation due to an increase in the signal-to-noise ratio, the object under investigation is exposed to a radio pulse for the allotted time of the study, in which the phases are one Ka half harmonic oscillations are set ON or 180 °, and the second half 90 or 270 in accordance with the coding Hadamard matrix, and the signals resulting from a double exposure. 35 45 five averaged. 13 12 l-i if f qsue.l In fi / L fV A L f v / -X- 4- f- 4-A # - 4-- - “n V l A l fieL . and 1.1 11 I Lj-.fli .. Y ЫIB If 19 and fi &. / -. ig7 Rih)) fpi / e.6 I {oJ} (J H- / Fig.8