Off Resonance & Eddy Current Artifact Correction
Click on to view the abstract pdf. Click on to view the recorded presentation.
Tuesday 8 May 2012
Room 212-213  10:00 - 12:00 Moderators: Clemens Bos, Pauline Worters

10:00 0215.   
Magnetic Resonance Encephalography Reconstruction with Magnetic Field Monitoring
Frederik Testud1, Jakob Assländer1, Christoph Barmet2, Thimo Hugger1, Benjamin Zahneisen1, Klaas Prüssmann2, Jürgen Hennig1, and Maxim Zaitsev1
1Medical Physics, University Medical Center Freiburg, Freiburg, Germany, 2Biomedical Engineering, University & ETH Zürich, Zürich, Switzerland

In the last years field probes have been used for dynamic magnetic field monitoring in order to reduce image artifacts due to imperfect gradient coils and amplifiers, eddy currents, B0 drifts and patient breathing by taking the measured field dynamics into account in the image reconstruction. Long Magnetic Resonance Encephalography experiments suffer from B0 and gradient drifts. These can be corrected for by interleaved measurement of the k-space trajectory and the imaging experiment. This is performed in the present work with 4 1H field probes combined with 28 channels of a 95 channel head coil.

10:12 0216.   
Higher-order monitoring of physiological field fluctuations in brain MRI at 7T
Signe Johanna Vannesjo1, Christoph Barmet1, Yolanda Duerst1, Simon Gross1, David O Brunner1, and Klaas P Pruessmann1
1Institute for Biomedical Engineering, University and ETH Zurich, Zurich, Switzerland

Field fluctuations caused by breathing and limb motion can have a strong degrading effect on quality of brain images at high fields. Partly, the resulting artifacts can be corrected for by proper f0-demodulation of the object signal, based on field data from navigators or concurrent field monitoring. In this work, we investigate the contribution of higher-order field components to the perturbations, and use concurrent field monitoring up to 2nd order to additionally improve image quality in high-resolution T2*-weighted GRE sequences.e 2nd-order concurrent field monitoring to correct for induced artifacts.

10:24 0217.   A New Type of Gradient: The Detection Frequency Gradient. A New Capability: Retrospective Shimming
Jonathan C Sharp1, Scott B King2, Mike Smith2, and Boguslaw Tomanek1
1Institute for Biodiagnostics (West), National Research Council of Canada, Calgary, Alberta, Canada, 2Institute for Biodiagnostics, National Research Council of Canada, Winnipeg, Manitoba, Canada

We introduce a new type of field gradient: the Detection Frequency Gradient (DFG), with detection frequency defined as lower case Greek omegaEMF - lower case Greek omegaPRECESSION. This gradient has some interesting properties, most notably the ability to adjust the shim retrospectively (post-acquisition). This represents genuine B0-shimming: signals from spins dephased by unwanted B0-gradients are brought back in to phase. This capability arises from a dynamically-defined B1 detector field, synthesized retrospectively by a time-dependent weighted combination of NMR signals from a receive coil array. Applications include post-acquisition shimming and eddy-current correction. The signal weightings are calculated from receiver coil field maps and a target B0-inhomogeneity.

10:36 0218.   
Automatic off-resonance correction with piecewise linear autofocus
Travis Smith1, and Krishna Nayak1
1Electrical Engineering, University of Southern California, Los Angeles, CA, United States

We present a new method to correct off-resonance blurring in spiral and radial images without knowledge of the field map. The image is divided into blocks and linear field map estimation and correction are performed on each block. The local linear coefficients are estimated through a combination of k-space spectral analysis and mapdrift, an image-domain correlation technique. The deblurring performance is comparable to field map-based techniques. The method does not use objective functions, requires only a blurry image (magnitude and phase) and a trajectory time map, and is suitable for low-SNR and fine-resolution images.

10:48 0219.   Novel automatic off-resonance correction without field maps in spiral imaging using L1 minimization
Hisamoto Moriguchi1, Kohki Yoshikawa2, Morio Shimada2, Shin-ichi Urayama3, Yutaka Imai1, Manabu Honda4, and Takashi Hanakawa4
1Radiology, Tokai University, Isehara, Kanagawa, Japan, 2Radiological Sciences, Komazawa University, Tokyo, Japan, 3Human Brain Research Center, Kyoto University, Kyoto, Japan, 4Functional Brain Research, National Center of Neurology and Psychiatry, Tokyo, Japan

A primary disadvantage of spiral imaging is blurring artifacts due to off-resonance effects. Most spiral off-resonance correction methods require a frequency field map. However, to acquire a field map increases the scan time. In this study, a novel spiral automatic off-resonance correction method without field maps has been demonstrated. In the newly proposed method, L1 minimization is used as a new criterion to determine the correct off-resonance frequencies. Frequency estimation process of the L1 min off-resonance correction is robust and the estimated field map shows reduced errors. The L1 min off-resonance correction has significant potential for faster spiral acquisition.

11:00 0220.   
Chemical species separation with simultaneous estimation of field map and T2* using a k-space formulation
Jose Luis Honorato1,2, Vicente Parot1,2, Cristian Tejos1,2, Sergio Uribe2,3, and Pablo Irarrazaval1,2
1Electrical Engineering, Pontificia Universidad Catolica de Chile, Santiago, Chile, 2Biomedical Imaging Center, Pontificia Universidad Catolica de Chile, Santiago, Chile, 3Department of Radiology, Pontificia Universidad Catolica de Chile, Santiago, Chile

In this abstract we present a novel technique for chemical species separation using a signal model formulated entirely in k-space. The model includes T2* decay, field inhomogeneity and a variable time map. By using a variable time map this method is able to separate species without artifacts produced by chemical shift, field inhomogeneity or T2* decay.

11:12 0221.   
Image Denoising Exploiting Sparsity and Low Rank Approximation (DSLR) in Slide Encoding for Metal Artifact Correction
Sangcheon Choi1, Hahnsung Kim2, and Jaeseok Park1
1Brain and Cognitive Engineering, Korea University, Seoul, Korea, Republic of, 2Electrical and Electronic Engineering, Yonsei University, Seoul, Korea, Republic of

Metal-induced field inhomogeneity is one of the major concerns in magnetic resonance imaging near metallic implants. Slice encoding for metal artifact correction (SEMAC) is an effective way to correct severe metal artifacts by employing additional z-phase encoding steps for each excited slice against metal-induced field inhomogeneity and view angle tilting (VAT). Despite the advantages of metal artifact correction, since noisy resolved pixels are included in image reconstruction, SEMAC suffers from noise amplification. SEMAC with noise reduction , which employs a two-step approach (rank-1 approximation along the coil dimension followed by soft thresholding in the slice direction), does not consider noise correlation of coils and results in a direct tradeoff between image accuracy and de-noising. Thus, to further expedite noise reduction in SEMAC, in this work we develop a novel image de-noising algorithm that exploits 1) low-rank approximation using strong correlation of pixels (x-z) in the slice direction (t), 2) Best Linear Unbiased Estimator (BLUE) image combination in the coil direction with noise correlation, and 3) recovery of distorted slice profile using the sparsity of signals in the slice direction with orthogonal matching pursuit (OMP).

11:24 0222.   Extracting Phase Contrast from MAVRIC Images Near Metal Implants
Kevin M Koch1, Matthew F Koff2, Weitian Chen3, and Hollis G Potter2
1Applied Science Laboratory, GE Healthcare, Milwaukee, WI, United States, 2Department of Radiology and Imaging, Hospital for Special Surgery, New York, NY, United States, 3Applied Science Laboratory, GE Healthcare, Menlo Park, CA, United States

The MAVRIC Multi-Spectral Imaging technique acquires 4D datasets, resolving spins around metal implants in space and Larmor frequency offset. Here, we demonstrate a novel processing technique that extracts phase-contrast images from conventional MAVRIC acquisitions. Dominant trends induced by metal hardware in the phased images are removed to reveal underlying tissue with phase-contrast. The presented method may offer a means to differentiate types of adverse tissue responses near metal implants.

11:36 0223.   
Fully-Refocused Spatiotemporally-Encoded MRI: Robust MR Imaging in the presence of metallic implants
Noam Ben-Eliezer1, Eddy Solomon2, Elad Harel3, Nava Nevo4, and Lucio Frydman2
1Center for Biomedical Imaging, New-York University, New-York, NY, United States, 2Chemical-Physics, Weizmann Institute of Science, Rehovot, Israel,3Chemistry, Northwestern University, Evanston, IL, United States, 4Biological-Regulation, Weizmann Institute of Science, Rehovot, Israel

A new MR encoding approach has been recently introduced based on sequential encoding of the image spatial domain. An interesting aspect of this Spatiotemporal-Encoding (SPEN) technique, stems from its ability to carry out a progressive, voxel-by-voxel refocusing of all T2* dephasing throughout the data acquisition, allowing it to overcome sizable field inhomogeneities. This work demonstrated SPEN’s potential for imaging near metallic implants, using in-vivo mouse models. Cartesian and Back-Projected SPEN MRI were implemented on a 7T microimaging unit, and compared versus conventional Spin-Echo scheme analogues. In all cases, unambiguously superior images arise from the fully refocused spatiotemporally-encoded protocols.

11:48 0224.   Effective and Flexible Eddy Current Compensation for Delta Relaxation Enhanced MR Imaging
Uvo Christoph Hoelscher1, and Peter Michael Jakob1,2
1Research Center for Magnetic Resonance Bavaria, Wuerzburg, Germany, 2Experimental Physics 5 (Biophysics), University of Wuerzburg, Wuerzburg, Germany

A new imaging method called dreMR uses a variable B0 field to generate a novel MR contrast. The variable B0 field induces eddy currents which substantially decrease the quality of dreMR images. The abstract presents a compensation method which can easily be implemented in the scanner software. The method can be used for any imaging sequence, does not require additional hardware and reduces the eddy current artifacts to a very small level.