JPH0271727A - Nuclear magnetic resonance imaging device - Google Patents

Abstract

PURPOSE:To simply prevent generation of an artifact by extracting and removing, from data, external noise which is mixed into a measurement signal during measurement of a nuclear magnetic resonance signal and which is discrete and impulsive. CONSTITUTION:Measured data in an MRI device is given by the formula (1), where S(u,v) is measured data, u is phase encode value, v is data sampling time, M(X,Y) is macro-magnetized distribution in the X-Y coordinates, X is a value on abscissa, and Y is value on coordinate. Further, the macro- magnetized distribution M(X,Y) on the X-Y coordinates which is an image obtained by the MRI device, is obtained by a two-dimensional Fourier transfor mation of the measured data S(u,v). Accordingly, if external noise which is discrete and impulsive is mixed into the measured data S(u,v), an arch fact on a slope, depending upon the strength and point of the mixing appears on the image. Accordingly, the component of the noise is extracted from the mea sured data S(u,v), and after the removal of the noise, the two-dimensional Fou rier conversion is carried so as to obtain an image having no artifact.

Description

Detailed Description of the Invention [Industrial Application Field] The present invention relates to a nuclear magnetic resonance imaging apparatus (hereinafter referred to as
When discontinuous and impulsive external noise, which occurs in the installation environment of the device (referred to as MHI device), mixes into the measurement signal, the MHI device prevents artifacts from occurring due to that noise.
This relates to an RI device.

[Conventional technology]

Conventional MRI equipment uses appropriate shielding means (depending on the installation environment of the equipment) to prevent external noise from entering measurement signals
It had a filter, a radio wave shield room, etc.

[Problem to be solved by the invention]

Conventional devices can measure and shield continuous and periodic noise from high-frequency facilities such as broadcasting stations. However, it can occur when other medical diagnostic equipment or fluorescent lamps installed around the equipment are turned on and off. For discontinuous and impulsive external noise, it is difficult to measure the noise intensity in the installation environment, and it is difficult to determine shielding performance, compared to the periodic noise described above.

Therefore, there is a problem in that it is impossible to shield against such impulsive external noise, and artifacts occur in images.

An object of the present invention is to easily prevent the generation of artifacts caused by the above-mentioned external noise without requiring a means for shielding against discontinuous and impulsive external noise, and therefore without requiring difficult determination of shielding performance. M who can
Our objective is to provide HI equipment.

[Means to solve the problem]

An MHI device in which a receiving system receives a nuclear magnetic resonance signal generated by irradiating a high-frequency pulse from a transmitting system to a subject to which a static magnetic field and a gradient magnetic field are applied, and obtains a reconstructed image based on the received nuclear magnetic resonance signal. This is achieved by extracting and removing discontinuous and impulsive external noise mixed into the measurement signal during the nuclear magnetic resonance signal measurement from the data obtained by the measurement.

[Effect]

Generally, the measurement data in an MRI device is expressed by the following formula (1)
It is two-dimensional data.

S(u,v)=f7M(X,Y)exp(i(uX+v
Y)) dXdY-=(1) Here, S (u, v): Measurement data U: Phase encode amount V; Data sampling time M (X, Y): Macroscopic magnetization distribution x on the X-Y coordinate plane:
x coordinate Y: Y coordinate At this time, macroscopic magnetization distribution M (X, Y) on the X-Y coordinate plane
is an image obtained by an MRI apparatus, and is obtained by performing two-dimensional Fourier transform on the measurement data S (u, v). Therefore, when discontinuous and impulsive external noise is mixed into the measurement data S (u, v), a diagonally lined artifact depending on the mixing position and intensity occurs in the image.

Therefore, by extracting and removing noise components from the measurement data S (u, v) and then performing two-dimensional Fourier transformation, an image without artifacts can be obtained.

In this way, the present invention extracts and removes noise components from the measurement data S (ut V) to obtain an image free of artifacts. Therefore, it does not require a shielding means, and therefore it is possible to solve the difficult problem of shielding performance. No determination is required, and the generation of artifacts caused by the above-mentioned external noise can be easily prevented.

〔Example〕

Hereinafter, the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing essential parts of an embodiment of an MRI apparatus according to the present invention, in which 1 is a receiving coil, 2 is a quadrature phase detector, and 3 and 4 are A/D converters. Reference numeral 5 denotes a computer, which includes a memory 5a for measurement data sR, a memory 5b for measurement data S1, and an image data memory 5c. 6 is the monitor, 7
is a discontinuous and impulsive extraneous noise.

In FIG. 1, a nuclear magnetic resonance signal from a subject (not shown) received by a receiving coil 1 is decomposed by a quadrature phase detector 2 into two low frequency components with a phase difference of 90 degrees.
The data are converted into digital data by A/D converters 3 and 4, respectively, and stored in memories 5a and 5b within the computer 5.

Here, the measurement data S (u, v) mentioned above was expressed as a complex number, but the two signal components whose phases differ by 90 degrees correspond to the real part and the imaginary part, respectively, and they are A/D converted. SR(u + V) +
It is called 8 + - (u + V). These measurement data SR+Sl are each subjected to high-speed two-dimensional Fourier transform by the computer 5, and further, the absolute value is determined by the following equation (2).

・・・・・・・・・(2) (F is Fourier transform) The data MA (X, Y) obtained by this is image data, which is displayed as a reconstructed image on the monitor 6 and used for diagnosis etc. be given

If the receiving coil 1 receives the external noise 7 at the same time as the nuclear magnetic resonance signal, the noise component will be mixed into the measurement data sR and SI, and artifacts will occur in the image. Therefore, in the apparatus of the present invention, the noise component is extracted and removed by the means described below.

FIG. 2 schematically shows a two-dimensional measurement data matrix in the above-mentioned apparatus of the present invention.

First, the measurement data SR and Sl obtained by one-phase detection and A/D conversion are positive and negative data, and the noise components have also been resolved, so the measurement data sR and S
The absolute value sA of , is determined by the following equation (3), and the data format is set to make it easy to determine the signal strength.

5A (u, v) = m mouth S + (u-v))' ---(
3) Here, the measurement data S^(U-1, v) before and after the phase encoding amount U of the measurement data S^(U, V) differs by
) and SA (un1.v) are generally values close to the measured data of 8 people (U, V). Therefore, a threshold value may be calculated from these measurement data of 8 people (u-1, v) and SA (un1.v), and the noise component mixed in the data value of the measurement data 5A (usV) may be extracted.

In this case, as shown in FIG. 3, the measurement data SA (u
+V) is a two-dimensional inverse Fourier transform of the image data MA (X, Y), so the signal strength is strong near the center of the measurement data matrix, that is, the low frequency part, and weak at the periphery. There is. Therefore, in order to distinguish the noise component from the signal component at a constant ratio at all points in the measurement data matrix, it is necessary to set a different threshold value depending on each data sampling time V. There,
The threshold value is successively changed from the left and right end points of the measurement data matrix, and the threshold value is set so that it becomes larger closer to the center.

In this embodiment, using such a threshold value, impulsive external noise is extracted and removed from the measurement data 5A (uv v) by the computer 5 serving as an image reconstruction means. The operation will be explained with reference to the flowchart of FIG. Here, we will take as an example a case where data from the left half of the u-th one-phase encoding interval, that is, the data from the first to the center of the data sampling time ■, is processed. First, the data sampling time ■ is set to 1, and the threshold value is set to the initial value given in advance (steps 100 and 101).
). Next, find the average value of the measurement data S^(u-1,v) and 8A(un1゜■) before and after the encoded amount U differs by one, and multiply this by the noise-to-signal discrimination ratio K given in advance. (Step 102). Next, this calculated value is compared with a threshold value (step 103), and if the calculated value is large, the calculated value is used as the threshold value (step 104). Next, compare the measurement data 5A (usV) and the threshold value (Step 105)
, when the measured data 5A (ulV) is larger than the threshold, the measured data 5A (un V) is regarded as data containing impulsive external noise, and the absolute value S^, the real part sR, and the imaginary part S! data is rewritten to 0 (step 106).

The above operation is repeated for the data from the first to the center of the data sampling time V (for the left half of the data in FIG. 3) (steps 107, 108).

Thereafter, the data from the last to the center of the data sampling time V (the right half of the data in FIG. 3) is processed in the same manner as described above. In addition.

In step 103, if the threshold value is greater than the calculated value, then in step 105;

If the measured data S^(u,v) is smaller than the threshold value, the process moves to step 107, respectively.

Hereinafter, by repeating these operations for the number of times of phase encoding, impulsive extraneous noise is extracted and removed from all measurement data.Based on the processed data, the computer 5 performs reconstruction calculations, and Get the image.

In the above explanation, in order to remove the noise component from the measurement data, we adopted a method of rewriting the measurement data containing noise to the data value O.
Data 5A (u-1, v) and SA (
un1. There is also a method of rewriting it to the average value of v). Also,
IEICE Journal D Vol, J71-D
There is also a method of replacing noise-containing data with estimated data using the algorithm described in "MRI image reconstruction algorithm from half-measured data using phase map" on pages 182 to 187, January 11988.

According to the above embodiment, the noise component can be extracted at a constant ratio to the measurement data at any point in the measurement data matrix, and the noise component can be removed at high speed.

〔Effect of the invention〕

According to the present invention, discontinuous and impulsive external noise mixed into the measurement signal during nuclear magnetic resonance signal measurement is extracted and removed from the measurement data, so there is no need for shielding means against the external noise. Therefore, it is possible to easily prevent the occurrence of artifacts caused by the above-mentioned external noise without requiring difficult determination of shielding performance.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the device of the present invention, and FIG.
The figure is a diagram showing an outline of a two-dimensional measurement data matrix in the same device as above; FIG. 4 is a flowchart for explaining the operation of the apparatus of the present invention shown in FIG. 1... Receiving coil, 2... Quadrature phase detector, 3, 4
... A/D converter, 5... Computer (image reconstruction means). 5a...memory for measurement data SR, 5b...memory for measurement data SI, 5C...image data memory, 6.
...Monitor, 7...Discontinuous and impulsive external noise. 75th edition number

Claims (1)

[Claims] 1. A receiving system receives a nuclear magnetic resonance signal generated by irradiating a high-frequency pulse from a transmitting system to a subject to which a static magnetic field and a gradient magnetic field are applied, and obtains a reconstructed image based on the received nuclear magnetic resonance signal. The nuclear magnetic resonance imaging apparatus includes image reconstruction means for extracting and removing discontinuous and impulsive extraneous noise mixed into the measurement signal during the measurement of the nuclear magnetic resonance signal from the data obtained by the measurement. A nuclear magnetic resonance imaging device featuring: