Electricity: measuring and testing – Particle precession resonance – Using a nuclear resonance spectrometer system
Reexamination Certificate
2000-12-29
2003-03-04
Lefkowitz, Edward (Department: 2862)
Electricity: measuring and testing
Particle precession resonance
Using a nuclear resonance spectrometer system
C324S322000, C324S318000
Reexamination Certificate
active
06529000
ABSTRACT:
BACKGROUND OF THE INVENTION
The field of the invention is nuclear magnetic resonance imaging (MRI) methods and systems. More particularly, the invention relates to a method and system for processing nuclear magnetic resonance (NMR) signals acquired during a scan in order to remove transient spike noise from the NMR signals and thereby eliminate artifacts produced by such noise in the reconstructed image.
When a substance such as human tissue is subjected to a uniform magnetic field (polarizing field B
0
), the individual magnetic moments of the spins in the tissue attempt to align with this polarizing field, but precess about it in random order at their characteristic Larmor frequency. If the substance, or tissue, is subjected to an additional magnetic field (excitation field B.) which is in the x-y plane and which is near the Larmor frequency, the net aligned moment, M
z
, may be rotated, or “tipped”, into the x-y plane to produce a net transverse magnetic moment M
t
. A radio-frequency (RF) signal, which is also denoted the nuclear magnetic resonance (NMR) signal, is emitted by the excited spins after the excitation signal B
1
is terminated, and this NMR signal may be received and processed to form an image.
MRI scanners include a large magnet assembly for producing the uniform polarizing field B
0
in a bore which is large enough to receive a patient. An RF coil surrounds the patient and is switched between a transmitter and receiver to produce the excitation field B, and to receive the resulting NMR signal. Additionally, three sets of gradient coils surround the RF coil to produce magnetic field gradients G
x
, G
y
and G
z
, and a shield is disposed therebetween to isolate the RF coil so that its uniform field is not disrupted. Typically, the region to be imaged is scanned by a sequence of measurement cycles in which the magnetic field gradients are switched on and off according to the particular localization method being used. The resulting set of received NMR signals are digitized and processed to reconstruct the image using one of many well known reconstruction techniques.
The NMR signals are very small and extraordinary measures are taken to shield the MRI system from external RF noise and to eliminate sources of internal noise. Nevertheless, generators of short-duration noise pulses persist and may elude location and elimination. These noise pulses are referred to as “spike noise”, “impulse noise” or “white pixels”, and lead to image artifacts with such vernacular names as corduroy and zebra artifacts. Sources of such noise include arcing due to partial discharges from intermittent electrical contacts or electrostatic discharge, and harmonics of fast transients such as those caused by ground loops. When such noise sources occur regularly, their source can be located and measures can be taken to eliminate them. This “hardening” process occurs at any new MRI installation, and eventually all the short-duration noise sources are eliminated except those which are intermittent and defy cost-effective diagnosis.
A number of strategies have been employed to mitigate the effects of intermittent noise sources. Such methods include the examination of the acquired NMR signals to locate noise spikes or the examination of the reconstructed image to locate the effects of such noise. These prior methods work when the noise spike occurs in NMR signals that are heavily phase or frequency encoded (i.e. on the edges of k-space), but they do not perform well when the noise spike occurs in NMR signals near the center of k-space. In the latter case the NMR signal magnitude is quite large and it is more difficult to discern signal from noise. Noise spikes detected by such methods are sometimes removed by interpolating between the adjacent values.
Another strategy, which is more effective in detecting and eliminating short-duration noise spikes near the center of k-space, was disclosed in U.S. Pat. No. 5,525,906 entitled “Detection and Elimination of Wide Bandwidth Noise in MRI Signals.” In that strategy, the NMR signal is processed by a noise filter or Transient Noise Suppression (TNS) system that includes a noise detector. The noise detector has a bandstop filter that is tuned to stop the NMR signals but to pass a range of frequencies outside the NMR imaging frequency band that include at least some of the spike noise. Because a considerable portion of the energy of short-duration spikes is located outside of the NMR imaging frequency band, the bandstop filter effectively isolates the spike noise from the NMR imaging frequency information. The bandstop filter thus provides an output signal that is an indication of the level of spike noise independent of the NMR imaging frequency information.
The magnitude of the output signal from the bandstop filter is then compared with a noise reference level at a comparator. When the magnitude of the output signal exceeds the noise reference level, a noise indication signal is produced (or is changed in its level) indicating that there is noise due to short-duration spikes. The noise indication signal can then be used to eliminate noise from the entire NMR signal by blanking out portions of the NMR signal whenever noise is detected, before the NMR signal is provided to an image reconstructor.
Although TNS systems are more effective at eliminating noise due to short-duration spikes near the center of k-space than the other systems mentioned above, TNS systems are highly frequency dependent. In particular, the stop band of the bandstop filter in a TNS system must be carefully set so that the filter passes the ranges of frequencies above and below the NMR imaging frequency band and not the NMR imaging frequency band itself. If the pass band of the bandstop filter encompasses the NMR imaging frequency band, the TNS system may mistake the high-magnitude signal components containing the imaging information for high-magnitude noise spikes, and inappropriately blank out portions of the NMR signal that contain useful information rather than noise. The high sensitivity of TNS systems to frequency is undesirable insofar as TNS systems must as a result be carefully and accurately implemented in order for the systems to properly remove noise due to short-duration spikes.
The high frequency sensitivity of TNS systems is also undesirable because it makes it necessary to configure a TNS system differently depending upon the frequency of operation of the MRI system (particularly the frequency of the polarizing field B
0
) in which it is implemented. Given the wide variety of MRI systems, and given that some MRI systems can operate at a variety of different frequencies, TNS systems must be repeatedly configured. Given that the tuning of TNS systems to MRI systems, and performance verification, can be costly, the high frequency sensitivity of TNS systems increases the cost of systems overall and places an undesirable constraint on the design of new MRI systems, particularly those that operate at multiple frequencies.
It would therefore be advantageous if a system could be developed for eliminating noise due to short-duration spikes from NMR signals and thereby mitigating the appearance of undesirable image artifacts from images created by MRI systems. It would particularly be advantageous if such a system could be developed that was successful in eliminating noise due to short-duration spikes even where the spikes were near the center of k-space. It would additionally be advantageous if such a system was not overly frequency sensitive in its operation, such that it could be easily implemented in a variety of MRI systems having a variety of frequencies of operation, or in MRI systems that operated at multiple frequencies of operation. It would further be advantageous if the system was low in cost and could be easily implemented.
BRIEF SUMMARY OF THE INVENTION
The present invention relates to, in a magnetic resonance imaging system, a method of processing a magnetic resonance signal including transient spike noise. The method includes receiving an initial signal related
Fetzner Tiffany A.
GE Medical Systems Global Technology Company LLC
Horton Carl
Lefkowitz Edward
Quarles & Brady LLP
LandOfFree
Method and system for processing magnetic resonance 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 and system for processing magnetic resonance signals..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for processing magnetic resonance signals... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3035191