Xue Feng¹, Sameul William Fielden¹, Hao Tan¹, and Craig H Meyer^1,2
1Biomedical Engineering, University of Virginia, Charlottesville, Virginia, United States, ²Radiology, University of Virginia, Charlottesville, Virginia, United States

Spiral-in/out bSSFP differs from spiral-out bSSFP in that it has a symmetric spiral-in gradient in front of a spiral-out gradient. This yields TE=TR/2 and leads to nulling of the 0th and 1st gradient moments via symmetry, which saves time compared to a flow-compensated spiral rewinder. However, eddy currents and other gradient imperfections of the gradient system affect the fidelity of the k-space trajectory and cause blurring and distortion in reconstructed images. In this abstract we compare cardiac spiral-in/out bSSFP images reconstructed using a k-space trajectory calculated using a single gradient delay model to those reconstructed using a model that incorporates gradient delays and eddy currents calibrated for each physical axis. We compare each of these to measured trajectories. In addition, we measured B0 eddy currents and analyzed their effect in spiral-in/out bSSFP imaging. Using estimated trajectory corrects most of the infidelity in k-space trajectory and improves the image quality. B0 eddy currents, however, have little effect on the reconstructed images.

Jochen Franke^1,2, Hank Donker³, Felix Mottaghy⁴, Christiane Kuhl³, Fabian Kiessling², and Volkmar Schulz^1,2
1Molecular Imaging Systems, Philips Research Europe, Aachen, North Rhine-Westphalia, Germany, ²Experimental Molecular Imaging, University of Aachen (RWTH), Aachen, North Rhine-Westphalia, Germany, ³Diagnostic and Interventional Radiology, University Hospital Aachen, Aachen, North Rhine-Westphalia, Germany, ⁴Nuclear Medicine, University Hospital Aachen, Aachen, North Rhine-Westphalia, Germany

The focus of this work is on the emerging methodology of magnetic resonance (MR)-based attenuation correction (MRAC) of positron emission tomography (PET) raw data. We propose a new MR sequence, comprising the functionality of ultrashort echotime (UTE) and the Dixon technique within a single examination. This UTE triple-echo (UTILE) sequence paired with a dedicated post-processing is capable to derive a truly MR-based attenuation map (µ-map) accounting for four compartments: soft tissue, adipose tissue, cortical bone and air.

Weiwei Zhang¹, and Yongchuan Lai^1
1GE Healthcare, Beijing, Beijing, China, People's Republic of

In this study, an accelerating phase correction method is introduced to reduce the pre-scan time for Fast Spin Echo pulse sequence (FSE). Compared with traditional phase correction method using an iterative approach with several shots data acquisition, this new method compresses all data acquisition within one shot.

Walter RT Witschey¹, Christian A. Cocosco¹, Daniel Gallichan¹, Gerrit Schultz¹, Hans Weber¹, Anna Masako Welz¹, Jürgen Hennig¹, and Maxim Zaitsev^1
1Medical Physics, University Medical Center Freiburg, Freiburg i. Breisgau, Germany

A method is presented to dynamically shim a single spatially encoded volume in steady-state sequences. Nonlinear phase preparation is used to temporally disperse gradient echoes originating from different positions within a volume. These echoes are locally shimmed using a k-space parcellation algorithm. Simulations and experiments are performed to eliminate bSSFP stop band artifacts along the direction of phase encoding.

Hing-Chiu Chang^1,2, Chun-Jung Juan³, Tzu-Chao Chuang⁴, and Hsiao-Wen Chung^2,3
1Global Applied Science Laboratory, GE Healthcare, Taipei, Taiwan, ²Institute of Biomedical Electronics and Bioinformatics, National Taiwan University, Taipei, Taiwan,³Department of Radiology, Tri-Service General Hospital, Taipei, Taiwan, ⁴Electrical Engineering, National Sun Yat-sen University, Kaohsiung, Taiwan

The PROPELLER-EPI has been shown useful for diffusion applications with reduced geometric distortion. A 2D phase correction with double-FOV reference image can be applied for oblique ghost reduction in each blade prior to PROPELLER-EPI reconstruction. We demonstrate the feasibility of an alternative method without the need for 2D fitting and determination of 2D margin, using double-FOV reference to estimate linear phase along two dimensions, and then combined with interlaced Fourier transform (FT) to reduce the oblique N/2 ghost. Its insensitivity to very low-resolution double-FOV reference is particularly suitable for RPOPELLER-EPI where low-resolution blades are combined to reconstruct high-resolution images.

Novena Rangwala^1,2, and Xiaohong Joe Zhou^2,3
1Department of Bioengineering, University of Illinois at Chicago, Chicago, Illinois, United States, ²Center for Magnetic Resonance Research, University of Illinois Medical Center, Chicago, Illinois, United States, ³Departments of Radiology, Neurosurgery and Bioengineering, University of Illinois Medical Center, Chicago, Illinois, United States

A technique is proposed to reduce ghosts arising from constant, linear, and “oblique” phase errors in EPI-PROPELLER with an arbitrary (oblique) imaging plane. Phase correction parameters for each blade of EPI-PROPELLER were calculated from three reference scans acquired along each of the three orthogonal axes. This technique was validated on a phantom and human volunteers. The results showed a reduction of the Nyquist ghosts by at least 80% in both long- and short-axis EPI-PROPELLER images.

Nii Okai Addy¹, Holden H Wu^1,2, and Dwight G Nishimura^1
1Electrical Engineering, Stanford University, Stanford, CA, United States, ²Cardiovascular Medicine, Stanford University, Stanford, CA, United States

Blipped echo planar imaging is very sensitive to timing errors resulting from MR gradient system imperfections causing ghosting artifacts to appear in images with no correction. Using a linear time-invariant model for the gradient system, trajectories achieved on the scanner can be estimated. This provides a general trajectory independent solution to correct for ghosting artifacts without an additional reference scan.

Nan-kuei Chen¹, Alexandru V Avram¹, and Allen W Song^1
1Brain Imaging and Analysis Center, Duke University Medical Center, Durham, NC, United States

Here we report an inherent and 2D phase correction technique to effectively remove EPI Nyquist artifacts, without needing any reference scan. A series of images are first generated by cycling through different possible values of 2D phase errors. An image with the lowest artifact level is then identified based on the background energy level. In comparison to traditional 1D correction methods, the developed 2D phase correction method is significantly more effective, particularly for oblique-plane EPI or in the presence of cross-term eddy current. The developed method can generally be applied to single-shot and segmented EPI, with or without parallel acceleration.

Yan Guo¹, Jiangbo Chen¹, and Xiaohua Jiang^1
1Department of Electrical Engineering, Tsinghua University, Beijing, China, People's Republic of

This work proposes an approach to simulate the stent artifacts in MRI based on electromagnetic field analysis. Both static and RF field distributions with a sample stent in a uniform imaging sample are calculated using the commercial FEM software JMAG 10.0 (JRI Solution, Limited, Japan). The images with stent artifacts are simulated by an MRI simulator based on the calculated field distributions.

Yin-Cheng Kris Huang¹, Chun-Jung Juan², and Te-Son Kuo^1
1Department of Electrical Engineering, National Taiwan University, Taipei City, Taiwan, ²Department of Radiology, Tri-Service General Hospital, Taipei City, Taiwan

Metallic implants caused image artifacts like in-plane distortion and T₂^*-related signal loss. As discovered in our preliminary studies that when dental braces were present, there were banding artifacts in the resulted TIDE bSSFP images, while the in-plane distortion was much less severe than the TSE images. The signal loss is likely due to the inherent off-resonance suppression mechanism for the findings of continuous phase drifts rather than random phases. Therefore, we applied system frequency adjustment to validate this hypothesis, and also used post-processing methods to combine the images acquired at various system frequencies, in hope to obtain an artifact-minimized image.

Sang-Young Zho¹, and Dong-Hyun Kim^1,2
1Electrical and Electronic Engineering, Yonsei University, Shinchon-dong, Seoul, Korea, Republic of, ²Radiology, Yonsei University College of Medicine, Shinchon-dong, Seoul, Korea, Republic of

View angle tilting technique showed great reduction of image distortion from B0 inhomogeneity. With slice-encoding, namely SEMAC, enable MR imaging near metal object. One drawback is long scan time and diffuculty of applying on the high-field due to increased frequency range. To mitigate this problem, spiral sequence is considered and result showed VAT is applicable even on metal object.

Hanne Wojtczyk¹, Petros Martirosian¹, Verena Ballweg¹, Hansjoerg Graf¹, and Fritz Schick^1
1Section on Experimental Radiology, University Hospital Tuebingen, Tuebingen, Baden-Wuerttemberg, Germany

In MRI near metals, electrical currents can be induced in the metals by RF pulses or by gradient switching (due to the imaging or concomitant gradient fields). The objective of the study was to systematically assess for the first time which effects both types of current have on SE phase images. A copper sheet was imaged in a coronal plane at 1.5 T using a SE sequence. The position of the sheet, phase encoding direction, bandwidth and transmit voltage were varied. For the measurements performed here, the SE phase images were dominated by effects from currents induced by gradient switching.

Jedrzej Burakiewicz¹, Christoph Kolbitsch¹, Geoffrey David Charles-Edwards^1,2, and Tobias Schaeffter^1
1Division of Imaging Sciences, King's College London, London, United Kingdom, ²Guy's & St Thomas' NHS Foundation Trust, London, United Kingdom

Rapid changes of heart rate create ghosting artefacts in ECG-triggered inversion recovery prepared MRI. A method is proposed to reduce the artefacts by real-time inversion time adaptation. The correction scheme was theoretically described, simulated, implemented on a scanner and tested using phantom and healthy volunteer brain scans. Variation in signal strength - the cause of artefacts - was reduced from 15% to 2.5%. This suggests a promising, easy to implement correction scheme increasing image quality by reducing artefacts.

Gigi Galiana¹, Jason Stockmann², Leo K. Tam², and Robert Todd Constable^1,2
1Diagnostic Radiology, Yale University, New Haven, CT, United States, ²Biomedical Engineering, Yale University, New Haven, CT, United States

This work examines the prototypical MR echo that would expected for a voxel of spins evolving in a strong quadratic (2z2-x2-y2) gradient field. This case is increasingly relevant given the growing interest in nonlinear imaging, and we report several surprising differences from the linear case, both in magnitude and phase of the echo. Furthermore, we show that neglecting these dynamics can lead to significant errors even in basic sequences, such as those routinely used for field mapping.

Sonya Bells¹, Sean Deoni^2,3, Ofer Pasternak⁴, and Derek K Jones^1
1CUBRIC, School of Psychology, Cardiff, United Kingdom, ²School of Engineering, Brown University, Providence, Rhode Island, United States, ³Centre of Neuroimaging Sciences-Institute of Psychiatry, King's College, London, United Kingdom, ⁴Brigham and Women's Hospital, Harvard Medical School, Bostan, MA, United States

Multi-component relaxometry of fast- and slow-T1 and T2 has previously been used to quantify aspects of tissue microstructure in brain tissue, notably – the myelin water fraction (MWF). The myelin water content is estimated by attributing the short T2 relaxation component to the water trapped within the myelin sheath. However, the amount of partial volume contamination from cerebral spinal fluid (CSF) is unknown and until now. Here, We investigate the effects of partial volume contamination due to CSF and propose a novel approach to correcting the problem of partial volume errors in mapping myelin water content.

Hendrik Laue¹, Dennis Doelschel², Felix Gremse², Matthias Günther¹, Fabian Kiessling², and Heinz-Otto Peitgen^1
1Fraunhofer MEVIS, Bremen, Bremen, Germany, ²Experimental Molecular Imaging, RWTH (University of Aachen), Aachen, Germany

Quantitative DCE-MRI requires fast sampling of the AIF. Fast aquisitions are prone to artifacts such as ghosting. We compared two methods correcting AIF measurements and compared the results on 10 Datasets. The first method removes affected timepoints by identifying increased signal outside the body. The second method splits the images and rejoins it so that the signal lost by the ghosting is recollected. Both methods yield an improvement. The first shows a strong improvement performs better but at the cost of about half of the time-points. The second is not as strong but conserves all time-points.