Digital Filter Banks in MR Measurement of Gradient Magnetic Fields

Abstract.The sensitivity of spectroscopic methods based on nuclear magnetic resonance (MR) is limited, in particular by the magnitude of noise in the signal being measured. In MR tomography and, above all, in localized spectroscopy and spectroscopic MR imaging, this problem becomes even more pronounced. When gradient magnetic fields are used, it cannot be fully ruled out that there will be a change in the basic magnetic field due to the eddy currents in conducting materials in the neighborhood of the sample being measured. This results in a local change in instantaneous frequency of the resonance of nuclei and in a distortion of spectral lines or MR image. For methods that eliminate this distortion and for an accurate calculation of the constants of (in particular long) preemphasis filters, techniques have been developed and experimentally tested that are based on measuring the instantaneous frequency of the signal detected with a very low signal-to-noise ratio. Adaptive filtering methods and filtering based on filter banks have been developed to reduce the level of noise. Results of these two types of filtering are described in the paper. The filtering techniques developed can be used also in other applications and thus contribute to increasing quality of methods for examining the properties of biological and chemical substances.

[1]  K. Bartusek,et al.  First-Data-Point Problem and the Baseline Distortion in Fourier-Transform NMR Spectroscopy with Simultaneous Sampling , 1994 .

[2]  A. Caprihan,et al.  A new method for flow velocity measurement: Frequency encoded NMR , 1986, Magnetic resonance in medicine.

[3]  J. Kaiser,et al.  Wavelet construction using Lagrange halfband filters , 1991 .

[4]  P. Vaidyanathan Multirate Systems And Filter Banks , 1992 .

[5]  P. G. Morris,et al.  Nuclear Magnetic Resonance Imaging in Medicine and Biology , 1986 .

[6]  K. Bartušek,et al.  MR measurement technique of rapidly switched gradient magnetic fields in MR tomography , 2005 .

[7]  L. Rabiner,et al.  FIR digital filter design techniques using weighted Chebyshev approximation , 1975, Proceedings of the IEEE.

[8]  R Gruetter,et al.  Localized eddy current compensation using quantitative field mapping. , 1998, Journal of magnetic resonance.

[9]  Michael H. Neumann,et al.  Exact Risk Analysis of Wavelet Regression , 1998 .

[10]  Rapid average‐flow velocity measurement by NMR , 1986, Magnetic resonance in medicine.

[11]  Philip E. Gill,et al.  Practical optimization , 1981 .

[12]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[13]  Karel Bartusek,et al.  Automatic Adjustment of Time-Variant Thresholds When Filtering Signals in MR Tomography , 2005, ICN.

[14]  Nouna Kettaneh,et al.  Statistical Modeling by Wavelets , 1999, Technometrics.

[15]  E. Gescheidtova,et al.  Magnetic resonance technique of gradient magnetic field measurement , 2003, Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (IEEE Cat. No.03CH37439).

[16]  Ingrid Daubechies,et al.  The wavelet transform, time-frequency localization and signal analysis , 1990, IEEE Trans. Inf. Theory.

[17]  A. Bruce,et al.  Understanding WaveShrink: Variance and bias estimation , 1996 .