JP3405720B2 - Magnetic resonance diagnostic equipment - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a so-called multi-voxel MRS compatible magnetic resonance diagnostic apparatus for acquiring a magnetic resonance spectrum for obtaining information such as a molecular structure of a target substance for each of a plurality of volume parts (hereinafter referred to as voxels). .

[0002]

2. Description of the Related Art Proton density images, T1-weighted images, and T2 are information that can be extracted from a subject using magnetic resonance.
The morphological information represented by an emphasized image is the mainstream.

On the other hand, MR spectroscopy (Magnetic Resonance Spec)
Roscopy) obtains a magnetic resonance spectrum in which the separation of the resonance frequency from a certain reference frequency is expressed in ppm, and from the magnetic resonance spectrum, the molecular structure of the target substance,
This method is mainly effective for diagnosing metabolic functions, because it can obtain information on the chemical environment, concentration, etc.

[0004] For example, when diagnosing the metabolism of amino acids in the brain, that is, the metabolism in which amino acids such as glutamic acid are synthesized after glucose, which is the main energy source in the brain, is taken into the brain, ¹³ C-labeled glucose is used. Then
It becomes possible to trace how ¹³ C is incorporated into amino acids such as glutamic acid.

¹³ C and ³¹ useful for such metabolic diagnosis
The sensitivity of P is lower than that of proton ( ¹ H) by three orders of magnitude or more. Therefore, in order to achieve the same sensitivity as proton, the voxel size is increased and signal processing is performed for each voxel. The voxel method is common.

In the multi-voxel method, the sampled MR signal is first subjected to Fourier transform processing, and the result is phase-corrected for each voxel.
Signal processing such as curve fitting, baseline correction, and frequency shift is performed.

The signal processing such as phase correction has a relatively large processing amount, and it has been a problem that the processing takes a long time.

[0008]

SUMMARY OF THE INVENTION An object of the present invention is to reduce the processing amount of signal processing such as phase correction and curve fitting in a magnetic resonance diagnostic apparatus compatible with multi-voxel MRS.

[0009]

[Means for Solving the Problems] A specific radionuclide in a local excitation region of a subject in a static magnetic field is excited by a high-frequency magnetic field and is saturated so that no signal is output from a predetermined saturation region. In the magnetic resonance diagnostic apparatus, which collects magnetic resonance signals generated from the specific nuclides in the region and processes the magnetic resonance signals by the multivoxels to obtain a frequency spectrum relating to the specific nuclides for each voxel, in the multivoxels. It has a function of performing signal processing only on some voxels of the above and displaying so that unprocessed voxels and voxels in which the signal processing has an error can be distinguished on the morphological image.

[0010]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, a device according to the present invention will be described with reference to the accompanying drawings in preferred embodiments.

FIG. 1 is a block diagram showing the configuration of a magnetic resonance diagnostic apparatus according to a preferred embodiment of the present invention. In the figure, the static magnetic field magnet 1 is for generating a static magnetic field in the imaging region and is provided with a superconducting or normal electric coil. A shim coil 2 is arranged inside the shim coil 2 in order to improve the spatial uniformity of the magnetic field strength of the static magnetic field. further,
A gradient coil 3 that generates a gradient magnetic field is arranged inside the shim coil 2. As is well known, the gradient magnetic field has the same direction as the static magnetic field, but the strength of the magnetic field has a gradient with respect to the spatial axis (three orthogonal axes), and this gradient magnetic field should be superposed on the static magnetic field. The spatial position information can be added to the magnetic resonance signal in the form of phase and frequency.

Further, inside the gradient coil 3, R
An F coil (high frequency magnetic field coil) 4 is arranged. This R
The F coil 4 converts a high frequency current supplied from the transmitting unit 7 into a high frequency magnetic field, is generally tunable, and can cope with various nuclides having different resonance frequencies. . The spin of magnetization excited by this high-frequency magnetic field recovers to the initial state with a time constant peculiar to the nuclide after the excitation is cut off. A magnetic resonance signal is generated during this relaxation process. This magnetic resonance signal is transmitted to the RF coil 4
It is received by the receiving unit 9 via the.

The receiver 9 amplifies a weak magnetic resonance signal,
Also detect. Although the RF coil 4 is shown and described here as being used for both transmission and reception, the RF coil 4 may be provided separately for the transmission only coil and the reception only coil. The magnetic resonance signal detected by the receiver 9 is sent to the data processor 15 via the data collector 11.

The sequence control unit 10 performs the high frequency magnetic field application, the gradient magnetic field application and the data collection according to the pulse sequence data corresponding to the multi-voxel MRS, the gradient coil power supply 5, the transmission unit 7, the reception unit 9 and the data collection. The part 11 is controlled.

The data processing unit 15 generates spectrum data for each voxel from the magnetic resonance signal by frequency analysis such as Fourier transform processing, and performs post-processing such as phase correction and curve fitting processing on the spectrum data. .

Here, in the MRS, post-processing such as phase correction and curve fitting processing for the spectrum data by the data processing unit 15 is indispensable, but in the present invention, the processing is targeted for all voxels of the multi-voxel. Is not performed, only some voxels in the multi-voxels are targeted, that is, for the spectral data of some voxels of the multi-voxels, post-processing such as phase correction and curve fitting processing is performed,
Post-processing is not performed on the spectrum data of the remaining voxels. This post-processing includes at least one of phase correction, curve fitting, baseline correction, and frequency shift.

A processing target voxel setting unit 12 is provided in order to set the voxels to be signal-processed in the data processing unit 15. As a method of setting the voxel to be processed, a method in which the operator manually specifies it via the console 13 and a voxel setting unit 12 to be processed does not involve the operator based on at least one of the local excitation region and the saturation region. We propose a method of automatically selecting a method and a semi-automatic method that uses both methods together. Any of these methods may be selectively made to correspond to the processing target voxel setting unit 12 in advance, or all of these methods may be made to correspond to the processing target voxel setting unit 12, and the operator Alternatively, it may be selectively used according to the instructions. The voxel selection work may be performed after the actual imaging (data sampling), or may be performed in the preparatory stage including so-called positioning for specifying the local excitation region and the saturation region before starting the data sampling. Good.

First, the manual method will be described. In order to assist the operator to specify which voxel is to be processed or which voxel is to be excluded from the processing target, as shown in FIG. The wireframe representing the framework, together with the frame showing the local excitation region and the frame showing the saturation region,
It is displayed graphically on the display 14. The operator looks at this frame and, for example, voxel numbers 27 and 2
The voxels of 8, 29, and 30 are designated as the processing target, or the voxels other than these are designated as the voxels which are not the processing target. As the designation method, the voxel number may be keyed in or may be designated by clicking the voxel on the screen using a pointing device. Furthermore, in order to facilitate the selection of voxels, these frames are set as shown in FIG.
It may be superimposed on a morphological image by MRI such as a T1-weighted image, a T2-weighted image, a density image, or the like captured in advance, or a morphological image captured by another modality such as X-ray CT.

Next, in the automatic method, the entire voxel is included in the local excitation region (voxel number 27).
-30, 35-38) is selected as a processing target. Further, at least a part of the voxels (voxel numbers 18 to 23, 26 to 31, 34) that are applied to the local excitation region.
~ 39, 42-47) may be selected as a processing target. Which one is selected is arbitrary by the operator. Furthermore, it is also possible to individually exclude voxels that have not been selected by the operator with respect to voxels that have been automatically selected by any method.

Further, the voxels ((voxel numbers 27 to 30, 3
5 to 38) or (voxel numbers 18 to 23, 26 to 3)
1, 34-39, 42-47)) some voxels
It is excluded from the processing target based on the saturation region. As a criterion for removing the voxel from the processing target, a voxel (voxel number 34) whose entire voxel is included in the saturation region may be excluded from the processing target, or even a part of voxels (voxel numbers 18, 26, 26) that are in the saturation region. 34, 42, 43, 39, 4
6, 47) may be excluded from the processing target. Which criterion is adopted is arbitrary by the operator.

In the semi-automatic method, the operator sets a region of interest in the local excitation region, and the voxel to be processed is automatically selected based on the region of interest. In this case as well, the voxel (voxel number 37) included in the region of interest may be selected as the processing target for the entire voxel, or at least a part of the voxel (voxel number 37) that lies in the region of interest may be selected. Voxel number 28-
30, 36 to 38, 44 to 46) may be selected as the processing target. Which one is selected is arbitrary by the operator. Furthermore, it is possible to add a voxel which is not selected by the operator in addition to the voxel selected by any method in units of voxels.

As described above, since the phase correction and the curve fitting processing are performed only on the minimum required number of voxels selected from the multi-voxels by various methods, it is possible to process all the multi-voxels rather than all the multi-voxels. The processing amount can be reduced.

Here, the processing target voxel setting unit 12 indicates to the operator that the voxels which have been subjected to the phase correction and the curve fitting processing are not processed although the data are temporarily collected. It also has a function to do. As the presentation method, for example, FIG.
As shown in, the wireframe representing the framework of the multi-voxel is graphically displayed on the display 14 together with the frame indicating the local excitation region and the frame indicating the saturation region, in which the range of unprocessed voxels is blinked. An example is that the display color is different from that of the processing voxel.

Further, a device having a function of not performing post-processing such as phase correction and curve fitting processing on voxels (error voxels) in which spectrum data shows an abnormality due to mixing of water signals and fat signals. In this case as well, the processing target voxel setting unit 12 has a function of presenting to the operator that the voxel is an error voxel. Similar to unprocessed voxels, the range of error voxels flashes or the display color is different from that of processed voxels. Furthermore, the blinking speed and display color of the error voxel
The two can be distinguished by making them different from those of the unprocessed voxels.

By displaying the unprocessed voxels or error voxels in this way, the operator can confirm them, and the operator can select one of the unprocessed voxels or the error voxels of interest as necessary. Alternatively, some voxels can be designated to perform post-processing. Of course, in order to facilitate the selection of voxels for which this post-processing is performed, these frames are pre-captured T1-weighted images,
It may be superimposed on a morphological image by MRI such as a T2-weighted image, a density image, or a morphological image captured by another modality such as X-ray CT.

The present invention is not limited to the above-described embodiment, and can be carried out in various modifications without departing from the scope of the invention in an implementation stage. Further, the above embodiment includes various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, some constituent elements may be deleted from all the constituent elements shown in the embodiment.

[0027]

According to the present invention, signal processing is performed only for necessary voxels, and the processing amount can be reduced as compared with the case where all multi-voxels are processed, and unprocessed voxels and The error voxels are displayed so that the operator can see them.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a magnetic resonance diagnostic apparatus according to a preferred embodiment of the present invention.

FIG. 2 is a diagram showing voxels set by a processing target voxel setting unit in FIG. 1;

FIG. 3 is a diagram showing voxels set by a processing target voxel setting unit in FIG. 1;

FIG. 4 is a diagram showing a display example of error voxels by the data processing unit of FIG. 1;

[Explanation of symbols]

1 ... Static magnetic field magnet, 2 ... Shim coil, 3 ... Gradient coil, 4 ... RF coil, 5 ... Gradient coil power supply, 6 ... Shim coil power supply, 7 ... Transmitter, 9 ... Receiver, 10 ... Sequence control unit, 11 ... Data collection unit, 12 ... Process target voxel setting unit, 13 ... Console, 14 ... Display, 15 ... Data processing unit.

Continuation of the front page (56) Reference JP-A-60-193447 (JP, A) JP-A-1-277545 (JP, A) JP-A-3-264049 (JP, A) JP-A-5-154131 (JP , A) JP 5-168606 (JP, A) JP 6-181910 (JP, A) JP 2000-163561 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) ) A61B 5/055 G01R 33/20

Claims (2)

(57) [Claims] 1. A high-frequency magnetic field excites a specific nuclide in a local excitation region of a subject in a static magnetic field, and saturates it so that no signal is output from a predetermined saturation region. In a magnetic resonance diagnostic apparatus for collecting a magnetic resonance signal generated from a specific nuclide and processing the magnetic resonance signal with a multi-voxel to obtain a frequency spectrum relating to the specific nuclide for each voxel, some of the multi-voxels A magnetic resonance diagnostic apparatus having a function of performing signal processing only on voxels and displaying unprocessed voxels and voxels in which signal processing has an error so that they can be distinguished on a morphological image. 2. The magnetic resonance diagnostic apparatus according to claim 1, wherein the signal processing includes at least one of phase correction, curve fitting, baseline correction, and frequency shift.