JPH0250730B2 - - Google Patents

Description

[Detailed Description of the Invention] [Technical Field of the Invention] The present invention relates to nuclear magnetic resonance (NMR).
magnetic resonans (hereinafter referred to as "NMR")
For example, diagnostic NMR-CT (CT:
In order to measure the density and relaxation time of symmetrical nuclei with high precision and high efficiency, NMR imaging devices such as computed tomograph (computerized tomograph) devices are used to measure the resonance frequency and excitation signal of target nuclides corresponding to the static magnetic field strength. This invention relates to an NMR imaging device that achieves high precision matching with a reference signal frequency.

[Technical background of the invention]

A diagnostic NMR-CT device is a device that obtains image information using NMR phenomena, and in particular, the obtained image information is used for diagnosis.

NMR, which generally performs measurements using NMR phenomena
In a device, high-accuracy measurement is achieved by matching the resonance frequency of the nuclide to be measured, which corresponds to the uniform static magnetic field strength applied to the specimen, with the frequency of the excitation signal, that is, the frequency of the reference signal used to obtain the excitation signal. required for.

In NMR devices used in conventional chemical analysis, the object to be measured is small and it is not possible to obtain two-dimensional images such as cross sections, so the required uniform magnetic field area is also small. Therefore, in this case, in order to maintain the relationship between the static magnetic field strength and the reference signal frequency in the so-called Bohr relationship (details will be described later), the detection probe is placed very close to the sample in the detection probe. In addition, a static magnetic field generator detects fluctuations in the static magnetic field using a resonance signal obtained from a nuclide with a resonance frequency different from that of the nuclide to be measured (this nuclide may be mixed into the sample) as a reference resonance signal. The so-called "NMR lock method" is used, which applies feedback to the

On the other hand, a diagnostic NMR-CT apparatus obtains two-dimensional images, and since the subject to be measured is an indivisible living body, the required uniform magnetic field area is large. This diagnostic NMR
When the above-mentioned NMR locking method is applied to a CT device, the required homogeneous magnetic field area is large, so special consideration must be given to the placement position of the reference nuclide. NMR-CT
This creates many problems in terms of device operability. Also,
Since the required uniform magnetic field area is large, it is not always possible to uniformly and accurately detect errors using the above reference nuclide. Therefore, a method that does not use such standard nuclides is desirable.

Furthermore, in the above magnetic field locking method, the uniform static magnetic field applied to the subject is variably controlled, and the strong magnetic field must be varied, making it difficult to perform stable control. In particular, when an air-core superconducting magnet is used in a static magnetic field generator, a magnetic field is generated in the superconducting coil in a permanent mode, that is, in a state where a so-called permanent current is passed through the coil that generates the magnetic field in a superconducting state. In this state, the magnitude of this current cannot be controlled externally, and the magnetic field lock method described above cannot be applied. Also,
In this case, it is difficult to maintain a perfect persistent current for a long period of time, and in this case, the current flowing through the superconducting coil gradually decreases, so it is especially necessary to take measures against the resulting fluctuations in the static magnetic field.

[Purpose of the invention]

The purpose of the present invention is to compensate for the deviation between the actual resonance frequency and the reference frequency that generates the excitation signal, without using a different nuclide from the measurement target, and without controlling the static magnetic field, with high precision and high performance. with efficiency
An object of the present invention is to provide an NMR imaging device that makes it possible to obtain NMR images.

[Summary of the invention]

The present invention excites the nuclear magnetic resonance of the nuclide to be measured under the application of a static magnetic field, performs frequency analysis on the resonance signal obtained from the object, determines the deviation between the resonance frequency and the reference signal frequency, and sends the result to the reference signal generator. It is characterized by being fed back as an error signal.

[Embodiments of the invention]

An embodiment of the present invention will be described in detail below based on its basic principle.

Theoretical analysis of NMR phenomena has been established by Bloch et al., and measurement methods can be roughly divided into continuous wave methods and pulse methods. Since the pulse method is common in diagnostic NMR-CT apparatuses, etc., the pulse method will be explained here.

When a rotating magnetic field with an angular frequency ω ₀ determined by the following equation derived from the Bohr relationship using a static magnetic field (strength) H ₀ is simultaneously applied to the target atomic nucleus at the same time as the static magnetic field H ₀ , it resonates with the rotating magnetic field and becomes Larmor's age. It exhibits a phenomenon called differential motion.

ω ₀ = γH ₀ …(1) (where ω ₀ : resonance angular frequency, H ₀ : static magnetic field strength,
γ: nuclear gyromagnetic ratio determined by the nuclide) This phenomenon is called the NMR phenomenon.

FIG. 1 shows the configuration of an embodiment in which the present invention is applied to a diagnostic NMR-CT apparatus. In addition, the first
The figure shows the configuration that is directly necessary for the present invention and is generally common to a diagnostic NMR-CT apparatus to which the present invention is applied, and the configuration that is not shown is necessary for a diagnostic NMR-CT apparatus. is not directly related to the present invention, and may differ depending on the various diagnostic NMR-CT apparatuses.

In FIG. 1, 1 is a static magnetic field generator that generates a uniform static magnetic field H ₀ for the imaging target area;
2 is an object to be examined (shown schematically), and 3 is a frequency-variable reference signal generator (in this case, a frequency-variable reference signal generating device that generates a reference signal with a reference angular frequency ωr corresponding to the resonance angular frequency ω ₀ of a symmetrical nuclide in a static magnetic field H _{0 )} . (hereinafter referred to as "SSG"), 4 is an excitation pulse generation unit that forms an excitation pulse _H1 consisting of a pulse-like high frequency (angular frequency ωr) signal based on the reference signal generated by the SSG 7, and amplifies the power; A transmitter coil that applies the excitation pulse _H1 generated by the excitation pulse generator 4 as a rotating magnetic field for NMR excitation to the subject 2, 6 a receiver coil that receives the NMR signal generated in the subject 2, and 7 a receiver coil. Received at 6
A low noise amplifier that amplifies the NMR signal; 8 a phase detection device that detects the phase of the amplified NMR signal using a reference signal with a reference angular frequency ωr generated by the SSG 3; 9
10 is a frequency analyzer that frequency-analyzes the NMR signal detected by the phase detector 8 to find the deviation Δω of the resonance angular frequency ω ₀ from the reference angular frequency ωr;
Feedback to SSG3, that is, according to the frequency deviation Δω
In addition to controlling the frequency of the SSG 3 and correcting the frequency deviation Δω, the SSG 3, excitation pulse generator 4, amplifier 7, phase detection device 8, and frequency analyzer 9
This is a system control device that controls the operation of the system. In this case, the transmitter coil 5 and the receiver coil 6 are separated, but they are integrated and the excitation pulse H ₁ is transmitted by a common coil.
In some cases, a so-called single-coil configuration is adopted in which the NMR signal is applied and the NMR signal is received.

In such a configuration, by applying a pulsed high-frequency rotating magnetic field determined by the above equation (1), that is, an excitation pulse _H1 , from the transmitter coil 5, the magnetic moment μ generated in the target atomic nucleus returns to an equilibrium state. The free induction decay (FID) signal induced at the time is set to the reference angular frequency ωr generated by SSG3.
When phase detection is performed by the phase detection device 8 using the reference signal of (This process of considering a situation is called ``viewing it in a rotating coordinate system.'')
Here, in general, the angle θ at which the magnetization M rotates as shown in FIG. 2 is expressed by the following equation in a rotating coordinate system of resonance frequency ω ₀ .

θ=γH ₁ t _P (2) (where t _P is the application time of H ₁ ) The detection system has the highest sensitivity when θ=π/2 (rad). In this example, exactly θ
Although it is not necessary to set = π/2, in the following
It is assumed that θ is close to π/2 to the extent that the sensitivity of the NMR-CT device is satisfied. Under such conditions, the FID signal V _FID detected by the phase detection device 8 at the reference angular frequency ωr is expressed by the following equation.

V _FID ∝m ₀ (x, y, z)e ^- { ^1/T2*+j( 〓 ₀ ^- 〓 ^r)t } =m ₀ (x, y, z)e ^-1/T2*・e ^{-j (} 〓 ₀ ^. 〓 ^r)t …(3
) (However, m ₀ (x, y, z): density of the target nuclide at (x, y, z), T ₂ ^* : spin of the target nuclide -
(Time constant due to spin relaxation time and magnetic field inhomogeneity) In equation (3), if the frequency deviation Δω is Δω=|ω ₀ −ωr|, the frequency deviation can be calculated by frequency analysis of V _FID with the frequency analyzer 9. It can be seen that the value of Δω can be known.

Here, consider the case where the frequency deviation Δω is fed back to the static magnetic field generator 1 as a current as in the conventional NMR lock type. From the above equation (1), the error ΔH ₀ of the static magnetic field strength H ₀ is ΔH ₀ =Δω/γ (4). Therefore, the current value i to be fed back is obtained from the following equation.

i=(ΔH ₀ )=(Δω/γ) …(5) (However, (H ₀ ): Static magnetic field strength of the normally conducting magnet H ₀
(relationship between current i and current i) In the conventional NMR lock method, since a sample nuclide different from the original target nuclide is used, it is possible to always detect an error signal, and the control by the feedback circuit is always performed. can be done. In the calculation using the above equation (5), errors may occur due to the shape of the magnet of the static magnetic field generator 1, the winding method of the coil, etc. Therefore, when detecting the frequency deviation Δω and feeding it back to the static magnetic field generator 1 as a current based on the above equation (5), a continuous feedback operation is required to avoid a decrease in accuracy due to the above error factors, and the above sample The presence of nuclides is important.
However, when considering the operability of the NMR-CT apparatus, the presence of the sample nuclide causes undesirable problems in terms of maintenance of the nuclide, positional relationship with the subject 2, and the like.

On the other hand, in this embodiment, the frequency deviation Δω is directly fed back to the SSG 3 as an error signal in order to ignore the error factor according to the above equation (5). This is the first feature of the present invention. In this way, the process according to the above equation (5) is avoided, and since no error factors are mixed in, it is possible to obtain sufficient accuracy without constantly detecting and feeding back error signals. Therefore, in this embodiment, the sample nuclide described above is not required, and the target from which the error signal is obtained is the diagnostic site of the subject 2 itself. This is the second feature of the present invention. Therefore, for example, it is sufficient to detect the frequency deviation Δω in a state where a static magnetic field H ₀ is applied just before imaging, and to correct the SSG3 accordingly, and there is no need to detect Δω during imaging, and the sample nuclide described above is not required. Therefore, the sequence time of the NMR-CT apparatus is not affected.

That is, in the configuration shown in FIG.
The NMR-CT apparatus is fixed at the angular frequency ω ₀ determined by the above equation (3). Whether the frequency deviation Δω is positive or negative can be easily determined by using, for example, a so-called (quadrature) quadrature detection (QD) as a detection method in the phase detection device 8.

Here, SSG3 will be explained in detail. In this embodiment, information on the frequency deviation Δω is used as an error signal.
Given to SSG3. Since this signal is the result of frequency analysis (which may be considered as fast Fourier transform), it is advantageous to treat it as a digital signal when considering the efficiency of the NMR-CT apparatus. Considering these points, SSG3 is a so-called PLL.
A configuration using a frequency synthesizer using a (phase locked loop) is suitable.

FIG. 3 shows the basic configuration of the SSG 3 using a PLL frequency synthesizer.

In FIG. 3, 31 is a highly stable oscillator that generates a PLL reference signal of frequency _c , 32 is a phase comparator,
33 is a low-pass filter as a loop filter; 34 is a voltage-controlled oscillator (hereinafter referred to as "VCO") that oscillates at a frequency of ₀ according to the input voltage _Vi ;
35 is a programmable frequency divider which divides an input signal of frequency ₀ by a frequency division ratio of 1/n according to input information Δω and inputs the frequency to the phase comparator 32.

To briefly explain the operation, the phase comparator 32 includes a highly stable oscillator 31 and a programmable frequency divider 35.
The output of is input, and the frequency of both these inputs i,
This closed circuit operates so that ₀ /n always matches. Therefore, the output of this circuit i.e. VCO
The relationship between the frequency ₀ of the output of 34 and the PLL reference signal frequency i is as shown in the following equation.

i= ₀ /n ∴ ₀ = n・i (6) In other words, the output frequency is determined by the division ratio 1/n of the programmable frequency divider 35 and the PLL reference signal frequency i.
₀ can be set freely. Also, output frequency ₀
The frequency stability of depends on the circuit constants of each part, but
The stability can be increased to almost the same level as that of the highly stable oscillator 31 that generates the PLL reference signal, and if a constant temperature crystal oscillator or the like is used as the oscillator 31, it can be used satisfactorily as an NMR-CT device. Therefore, the frequency division set value n of the programmable frequency divider 35 is changed by the digital error signal corresponding to the frequency deviation Δω which is the result of the frequency analysis, and a frequency corresponding to the fluctuation of the static magnetic field strength H ₀ is outputted.

The configuration shown in FIG. 3 is the most basic one, and although there are slight variations in the configuration due to frequency variable steps and frequency stability requirements, the configuration using a PLL synthesizer is basically the same. In this case, the frequency variable step of SSG3 needs to have the same level of resolution as the frequency analysis. Furthermore, considering the efficiency of signal transmission, it is considered appropriate that the output waveform be a sine wave.

Next, how to obtain the frequency deviation Δω will be explained.

FIG. 4 shows a timing sequence for determining the frequency deviation Δω.

As shown in Fig. 4, if a 90° pulse (high-frequency excitation pulse such that θ = π/ ₂ ) is applied as the excitation pulse H 1 to the subject 2 in the static magnetic field H ₀ , FID ₁ or FID FID signals like ₂ can be observed. FID ₁ is an example of the FID signal when ωr≠ω ₀ , and FID ₂ is an example of the FID signal when ωr=ω ₀ .
The frequency spectra of these FID signals FID ₁ and FID ₂ are shown in Figure 5, respectively.
In the frequency spectrum of FID ₁ , a spectrum appears at a position shifted from the origin by a frequency deviation Δω, and in the frequency spectrum of FID ₂ , the spectrum position coincides with the origin. From this, the principle of this embodiment described above can be easily understood.

When detecting FID signals, avoid errors caused by saturation of the detection system amplifier due to 90° pulses, improve S/N (signal-to-noise ratio) by averaging, and avoid the effects of signal attenuation due to magnetic field inhomogeneity. For this purpose, many NMR devices measure signals called NMR echoes. Echo signals can also be used in this embodiment. The timing sequence in this case is shown in FIG.

As shown in Fig. 6, a 90° pulse is applied to the object 2 in a static magnetic field H ₀ , and then a 180° pulse (θ=π
By applying a high-frequency excitation pulse (such that
can be determined. In this case, the S/N is improved by averaging the first echo and the second echo.

In addition, in FIGS. 4 and 6, a series of sequences starting from time ta is the first sequence, and time t _b
If we make the sequence starting from the second sequence,
If sufficient S/N is not obtained in the first sequence, the S/N is improved by sequentially repeating and averaging the second and subsequent sequences.
The accuracy of frequency deviation Δω verification and frequency control is improved.

Note that whether to detect the FID signal or the echo signal depends on the S/N of each method and the above (2), as in the case of the general pulse method.
If the deviation of θ from π/2 in the formula is Δθ,
When using a 180° pulse, the error should be determined taking into consideration the fact that the error will be 3Δθ, and the present invention is not particularly limited thereto.

Next, the static magnetic field generator 1 will be explained in detail.

Of course, an air-core normal conducting magnet may be used as the static magnetic field generator 1, but it is also effective to use an air-core superconducting magnet.

When using an air-core superconducting magnet, as mentioned earlier, while operating in permanent mode (persistent current state), if you want to control the current, you must cancel the permanent mode and adjust the current. , it is not desirable to adjust the current value using an error signal, since the characteristics of the air-core superconducting magnet cannot be utilized. Another method for controlling the static magnetic field strength H ₀ when an air-core superconducting magnet is used in the static magnetic field generator 1 is to wind a compensation coil around the air-core superconducting magnet and adjust the current value of the coil. It is also possible to do something like this. However, in this case, a compensation coil and a constant current source for driving the coil are separately required, which complicates the configuration.

On the other hand, if the present embodiment is used, the above-mentioned external device is not required, and the correction control sequence can be exactly the same as when using an air-core normal-conducting magnet. However, it is possible to further improve the accuracy of imaging and thus the accuracy of diagnosis without complicating the configuration of the device.

The effect of using an air-core superconducting magnet as described above can be roughly explained as follows.

Considering a superconducting magnet that can generate a magnetic field of 5000 Gauss, the change in magnetic field in permanent mode is set to 0.1 ppm/h, and the operating time of the NMR-CT device is set to 8 h/day.
Then, the fluctuation of the magnetic field is 4 mGauss/day.
Assuming that the nuclide to be measured is protons, the frequency fluctuation will be approximately 17.0Hz/day, and if the resolution of frequency analysis is 20.0Hz, the permanent mode will be canceled every day if the adjustment is made by canceling the permanent mode of the superconducting magnet. The current value must be adjusted by On the other hand, when this embodiment method is applied, the half width of the tuning section in the signal amplification section is 141.9kHz (Q = 150).
Assuming that the allowable variable range is approximately 71kHz, (71×
10 ³ / 17 =) It is not necessary to cancel the permanent mode for about 4176 days. In this case, when considering the spread of the frequency spectrum due to the pulse width of the excitation pulse H ₁ and the effects of evaporation of the coolant, the above-mentioned period during which the permanent mode does not need to be canceled is considered to be shortened. However, even if these effects are taken into account, if the method of this embodiment is applied to a superconducting magnet, one of the characteristics of a superconducting magnet, ``during permanent mode operation, the static magnetic field strength is stable and the current value is compensated.'' It is clear that the advantage of "no need for" can be fully utilized.

As mentioned above, the conventional NMR lock method
When applied to an NMR-CT device, it is necessary to provide a sample nuclide other than the measurement target at a location different from the subject, which limits operability. Also,
A method in which the signal from the measurement target (subject) is returned to the static magnetic field generator as an error signal using some method improves operability, but feedback cannot always be applied, so it depends on the error factors involved in the adjustment. An error will occur. Furthermore, when an air-core superconducting magnet is used in a static magnetic field generator, it is difficult to adjust the magnetic field without separately providing a compensation coil.

In contrast, according to this embodiment, the NMR signal from the subject (measurement target) is frequency-analyzed at a convenient timing, such as immediately before imaging, and the result is directly fed back to the SSG as an error signal. The SSG frequency can be made to accurately follow fluctuations in the static magnetic field strength, etc. Therefore, NMR-CT
It is possible to reduce the causes of errors without impairing the operability of the device, and it is possible to improve the imaging accuracy and, in turn, the diagnostic accuracy. Furthermore, the control sequences of conventional NMR-CT devices can be fully utilized and used without major changes. Furthermore, when using superconducting magnets in the NMR-CT system, SSG
Since static magnetic field fluctuations are compensated for, there is no need to cancel the permanent mode, and there is no need to take the trouble to provide an external device such as a compensation coil.

It should be noted that the present invention is not limited to the embodiments described above and shown in the drawings, but can be implemented with various modifications without changing the gist thereof.

For example, frequency analyzer 9, system controller 1
The functions such as 0 may be realized by software processing in an electronic computer for image reconstruction.

Furthermore, by superimposing a gradient magnetic field on the static magnetic field, a region in the static magnetic field may be limited and frequency deviation in a specific region may be compensated for.

In addition, the timing for performing correction control is also different from the original timing.
It may be performed not only before taking an NMR image but also after taking an image or during taking an image as necessary.

〔Effect of the invention〕

According to the present invention, it is possible to compensate for the deviation between the resonance frequency and the reference frequency due to static magnetic field fluctuations, etc., without using a different nuclide from the measurement target, and without controlling the static magnetic field, and to perform NMR imaging with high precision and high efficiency. We can provide NMR imaging equipment that can be obtained.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, FIG. 2 is a diagram showing the movement of the magnetization vector due to the excitation pulse H ₁ in a rotating coordinate system for explaining the embodiment, and FIG. A block diagram showing an example of the configuration when a PLL synthesizer method is adopted as the reference signal generation device in the same embodiment, and FIG. 4 shows an example of the excitation-Δω detection sequence in the same embodiment (when using an FID signal). 5 is a timing chart showing an example of the frequency spectrum in the same example, and FIG. 6 is a timing chart showing another example of the excitation-Δω detection sequence in the same example (when using an echo signal). be. DESCRIPTION OF SYMBOLS 1...Static magnetic field generator, 3...Standard signal generator (SSG), 4...Excitation pulse generator, 5...Transmitter coil, 6...Receiver coil, 7
...Low noise amplifier, 8...Phase detection device, 9...
Frequency analysis device, 10...system control device.

Claims (1)

[Claims] 1. Produce a nuclear magnetic resonance phenomenon in a specimen, detect the induced nuclear magnetic resonance signal, and obtain image information reflecting at least one of the spin density and relaxation time constant of a specific atomic nucleus in the specimen. In the nuclear magnetic resonance imaging apparatus to be obtained, an air-core superconducting magnet that generates a uniform static magnetic field substantially over the entire imaging region to be applied to the object, and a reference signal having a frequency corresponding to the target atomic nuclide, and a reference signal generator capable of adjusting and controlling the generated frequency; and an excitation signal applying device that forms a high-frequency excitation signal for nuclear magnetic resonance excitation based on the reference signal output of the reference signal generator and applies it to a subject. , a receiving and detecting device that receives and detects nuclear magnetic resonance signals induced in a subject;
a first control means for operating the air-core superconducting magnet and energizing the excitation signal applying device; and a first control means for frequency-analyzing the nuclear magnetic resonance signal detected by the reception and detection device; a frequency analysis means for determining a frequency deviation between the reference frequency and the resonant frequency when the excitation signal is applied to a subject in a uniform static magnetic field, and generating the reference signal in accordance with the frequency deviation obtained by the means; 1. A nuclear magnetic resonance imaging apparatus comprising: second control means for controlling the frequency deviation and correcting the frequency deviation.