Electricity: measuring and testing – Particle precession resonance – Using a nuclear resonance spectrometer system
Reexamination Certificate
1999-07-28
2002-04-23
Williams, Hezron (Department: 2862)
Electricity: measuring and testing
Particle precession resonance
Using a nuclear resonance spectrometer system
C324S309000
Reexamination Certificate
active
06377043
ABSTRACT:
BACKGROUND OF THE INVENTION
MR methods (MR=magnetic resonance) require exact knowledge of the temporal variation of the magnetic field, notably of its spatially linearly varying component (gradient field). If this variation is not known, artefacts are liable to occur in M images upon reconstruction, and spatially selective RF pulses could excite the nuclear magnetization in a region other than the intended region.
For conventional MR methods it is assumed that the magnetic field, notably the gradient field, exhibits the desired temporal behavior which is imposed by a waveform generator. However, this is the case only if no eddy currents occur in the gradient coils required to generate the gradient fields and if the gradient amplifiers, delivering the currents for the gradient coils, are exactly linear.
In order to adapt the temporal variation of the gradient field to the desired variation despite eddy current losses, it is known from EP-B 228 056 to apply the gradient waveform or pulse shape to the gradient amplifier via a pre-emphasis filter having a transfer behavior which is the inverse of that of the gradient coil. The transfer behavior of the gradient coil is determined, during a preceding measuring operation, on the basis of its pulse response and the filter parameters are chosen in dependence on the measured values. However, this method can be effectively applied only if the entire chain of electronic components of the gradient channel can be considered to constitute a linear and temporally invariant system. Non-linear characteristics of the gradient amplifiers and temperature-dependent gain factors, however, oppose these requirements.
In a method which is known from Proc. ISMRM 1996, 1405, the currents in the gradient coils are measured and it is assumed that the temporal variation of the gradient field corresponds to the temporal variation of the measured current. However, in that case eddy current effects are not taken into account. According to another known method (SMR 94, 484) the magnetic field is measured directly in the magnet and the k-space trajectories are determined therefrom. It is a major drawback of such a method, however, that such a measurement must be performed separately for each MR sequence used. This-is a very time-consuming operation which precludes, inter alia, the changing at will of the MR pulse sequence during the data acquisition (for example, interactive MR).
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a method enabling simple and accurate determination of the temporal variation of the magnetic field, notably of the gradient field.
This object is achieved by means of an MR method which includes the following steps:
measurement of the temporal variation of the current in or the voltage across a gradient coil,
determination of the temporal variation of the magnetic field, generated by the gradient coil, from the measured temporal variation of the current (or of the voltage) and from the pulse response of the gradient coil, or from a quantity derived therefrom.
The invention is based on the recognition of the fact that the gradient coils themselves constitute linear systems exhibiting a temporally constant transfer behavior. Therefore, it suffices to measure the pulse responses of these systems only once; during later examinations merely the temporal variation of the current or of the voltage is measured and the temporal variation of the magnetic field is calculated from this variation and from the stored pulse response, for example by convolution of the current with the stored pulse response. This enables accurate determination of the temporal variation of the magnetic field. It is to be noted that no requirements whatsoever are imposed as regards the linearity of the gradient amplifier (therefore, use can be made of inexpensive amplifiers exhibiting given non-linearities). It is not essential either that the eddy current effects are particularly small, because they are taken into account in the pulse response which is measured only once.
In a further embodiment, the invention includes utilizing the more exact knowledge of the temporal variation of the spatially linearly varying component of the magnetic field (also referred to as gradient field hereinafter), as acquired according to the invention, so as to enhance the reconstruction. In customary methods the reconstruction is based on k-space trajectories which result from the desired variation of the gradient fields or from only one current measurement without taking into account the eddy currents. However, if the actual k-space trajectories do not correspond to these values, image artefacts occur. According to the invention the reconstruction is based instead on the actual k-space trajectories obtained by measurement and calculation. The k-space is then no longer scanned along the ideal k-space trajectories, but various methods (such as the so-called gridding method) also enable perfect reconstruction in such cases. No preparatory measurements which are dependent on the relevant MR sequence are required for this purpose.
Gradient coils having a less favorable construction are also capable of generating, in addition to the desired, spatially linearly varying components of the magnetic field, higher-order components which spatially vary as a function of the second or a higher power. Such components may also cause image defects. However, use of reconstruction methods which also take into account such components for reconstruction, can lead to improved reconstruction.
In a further embodiment, taking account of the temporal variation of the magnetic field enables the design of RF pulses which excite the nuclear magnetization in predetermined spatial regions only.
Claim
5
describes an MR device for carrying out the MR method the device including a main field magnet, gradient coil system, a memory for storing the pulse response of a quantity derived therefrom of each gradient coil, a receiver for receiving MR signals, and a reconstruction unit with at least one measuring unit for measuring the current in or the voltage across each gradient coil during an MR examination and with means for determining the spatial and temporal variation of the magnet field of the gradient coils from the measured values and the stored pulse sequences.
REFERENCES:
patent: 4698591 (1987-10-01), Glover et al.
patent: 4950994 (1990-08-01), Glover et al.
patent: 0 228 056 (1986-12-01), None
patent: 0 395 248 (1989-04-01), None
Aldefeld Bernd
Holz Dietrich
Schomberg Hermann
Shrivastav Brij B.
US Philips Corporation
LandOfFree
Magnetic resonance method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Magnetic resonance method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Magnetic resonance method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2925768