Electricity: measuring and testing – Particle precession resonance – Using a nuclear resonance spectrometer system
Reexamination Certificate
2001-05-07
2003-04-22
Arana, Louis (Department: 2862)
Electricity: measuring and testing
Particle precession resonance
Using a nuclear resonance spectrometer system
C324S307000
Reexamination Certificate
active
06552541
ABSTRACT:
This application claims Paris Convention priority of DE 100 32 345.6 filed Jul. 4, 2000 the entire disclosure of which is hereby incorporated by reference.
BACKROUND OF THE INVENTION
The invention concerns a method of correcting disturbing influences on the MR (=magnetic resonance) signals of a substance disposed in the measuring volume of an MR apparatus excited by one or several RF (=radio frequency) excitation pulses, wherein an RF excitation pulse is irradiated onto the substance and a time-dependent MR signal produced thereby is detected and digitized in a phase-sensitive fashion.
A method of this type is disclosed e.g. by DT 23 52 315 C2 and EP 0 522 191 A1.
MR field stabilization should be as accurate as possible to keep the magnetic field as constant as possible. Towards this end, both small interfering amplitudes which produce small frequency changes as well as relatively large external magnetic interfering amplitudes should be compensated. Field instabilities have a disturbing effect, in particular, for prolonged experiments which can take several hours. Even with the most modern superconducting magnets, the magnetic field changes with time due to a very small but important electrical resistance of the coil itself and due to external influences, such as the surrounding temperature.
To compensate for field inhomogeneities, an MR signal is conventionally used as a reference in a control circuit, wherein the shape of the dispersion signal is utilized. The amplitude in the center of the signal is zero and that amplitude is non-zero on both sides of the center. Most importantly, the sign of the signal changes during passage through resonance. If the initially detected signal is detected exactly on resonance and the magnetic field then changes, a signal is produced in a detector. The sign of the signal indicates the direction of change of the magnetic field. In a superconducting magnet, the so-called lock signal is amplified and fed to a coil which produces a magnetic field to either increase or decrease the main magnetic field. In an electromagnet, the lock signal controls the current supplied to the magnet. This method is relatively inaccurate since only a small portion of the signal is used to control the magnetic field. Exact determination of the passage of the signal through resonance is also difficult.
DT 23 52 315 C2 discloses a method of stabilizing the relationship between measuring frequency and magnetic field strength in a spin resonance spectrometer. The spin resonances of a standard substance introduced into the magnetic field of the spin resonance spectrometer are excited by RF pulses having a dense frequency spectrum. The frequency of the resonance signal of the standard substance, contained in the pulse response signal, is determined and compared with a reference frequency. The determined deviation provides a controlled variable for adjusting the magnetic field strength or the measuring frequency, with the frequency of the resonance signal of the standard substance being determined through Fourier analysis of the pulse response signal. The spin resonances of the standard substance are adjusted, after sufficient approximation of the relationship between measuring frequency and magnetic field strength, to the desired relationship using a continuous-wave signal to produce a suitable resonance signal for the standard substance. Such a pulsed control is very slow and therefore inaccurate.
EP 0 522 191 A1 discloses a method of compensating time-dependent field disturbances in magnetic fields of electromagnets having high field homogeneity. In particular, a dispersion nuclear signal u
x
of a comparative substance, located in the sample region of the superconducting electromagnet within which the magnetic resonance is measured, is recorded and used for compensation by producing a current in a field correction coil of the electromagnet in dependence on the dispersion signal. The absorption signal u
y
of the comparative substance is also recorded and compensation is carried out in dependence on the variable u
x
/u
y
and/or (1/u
y
) (du
x
/dt). This prior art method is also too slow for field stabilization required for a prolonged experiment.
It is therefore the underlying purpose of the present invention to realize a method for faster and more exact control, i.e. a method which decreases the sampling time and increases control sensitivity.
SUMMARY OF THE INVENTION
This object is achieved in accordance with the invention in a simple but effective fashion in that a time dependence s
i
(t
i
) of the detected and digitized MR signal is used to determine a time dependence &Dgr;&phgr;
i
(t
i
) of the phase of the MR signal relative to a predetermined reference phase &phgr;
refi
(t
i
) of a reference signal s
refi
(t
i
). The determined phase is then digitized and one or more correction or control values are determined therefrom.
In the inventive method, the phase dependence is the actually relevant variable for a deviation of the magnetic field from the desired value. In prior art, most of the measuring points play no or only a subordinate role in determining the magnetic field deviation. However, the values of a free induction decay (FID) at later times of the FID are of particular interest for control and correction. Advantageously, in the inventive method, nearly all measuring points are used for determination of the magnetic field deviation and thus for control. This guarantees considerably improved control accuracy and furthermore permits faster repetition with less measuring points for equal or improved control quality. Control is possible with means which are already present in an MR apparatus. Additional electronic equipment for controlling the frequency or the magnetic field is not required.
In a particularly preferred variant, the average slope of the phase of the MR control signal, relative to the reference phase, is determined by means of a linear regression through the corresponding time function &Dgr;&phgr;
i
(t
i
). The use of a straight line permits estimation of a previous frequency deviation, wherein the influence of noise on the estimate is suppressed by the straight line fit.
Another variant is characterized in that a curvature of the time function &Dgr;&phgr;
i
(t
i
) is determined and used for predicting the probable behavior of the system with respect to nominal deviations in the immediate future. Determination of the curvature further improves the control or correction since extrapolation permits prediction of the deviation from the ideal case, thereby producing adaptive and more accurate control.
In a further development, the curvature of the time function &Dgr;&phgr;
i
(t
i
) is determined by means of a polynomial fit of higher order. A polynomial fit permits particularly accurate approximation of the time function. Accurate determination of the actual frequency is thus possible. This is particularly important at the end of an FID.
A further method variant is characterized in that the contributions to the correction or controlled variable at different times t
i
from individual measuring points of the phase &Dgr;&phgr;
i
, relative to the reference phase, are weighted in dependence on the signal-to-noise-ratio. Stronger weighting of low-noise increases the control accuracy. It is, however, also feasible to weight in dependence on the signal amplitude. Signals having a large amplitude are thereby used to a larger extent for determination of the controlled or correction variable than are signals having smaller amplitudes.
In a further method variant, RF excitation pulses are irradiated onto the substance in the measuring volume at pulse angles of &agr;<<90°. This approach does not require waiting for establishment of an equilibrium to permit rapid determination of the controlled variables and high repetition rates, thereby increasing the speed of the method.
If the frequency of the reference signal is chosen to be different than the desired frequency and if this preferably constant deviation is taken into consideration for de
Arana Louis
Bruker Biospin MRI GmbH
Vincent Paul
LandOfFree
Method of correcting disturbing influences on the MR signals... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of correcting disturbing influences on the MR signals..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of correcting disturbing influences on the MR signals... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3080724