Kazim Gumus¹, Benedikt Poser¹, Brian Keating¹, Brian Armstrong², Julian Maclaren³, Thomas Prieto⁴, Oliver Speck⁵, Maxim Zaitsev⁶, and Thomas Ernst^1
1John A. Burns School of Medicine, U. of Hawaii, Honolulu, HI, United States, ²Dept. of Electrical Engineering and Computer, U. of Wisconsin-Milwaukee, Milwaukee, WI, United States, ³Dept. of Radiology, University Medical Center Freiburg, Freiburg, Germany, ⁴Neurology, Medical College of Wisconsin, Wauwatosa, WI, United States, ⁵Dept. Biomedical Magnetic Resonance, Otto-von-Guericke-University, Magdeburg, Germany, ⁶Dept. of Diagnostic Radiology, University Hospital Freiburg, Freiburg, Germany

Intra-scan head motion causes signal dropouts in DWI. A new method is presented to eliminate such artifacts. We used a Moiré-Phase-Target based tracking system to measure head motion between excitation and acquisition. Knowing also the timing and amplitudes of gradients, we determined the motion-induced residual gradient moment (M) and restored the gradient balance prior to readout by applying a blip gradient of moment -M. The method was tested on two volunteers who performed intentional head movements. Gradient moment correction successfully eliminated signal dropouts compared to scans without correction. This method should be feasible on most modern scanner platforms.

Himanshu Bhat¹, M. Dylan Tisdall², Andre Jan Willem van der Kouwe², Thorsten Feiweier³, and Keith Heberlein^1
1Siemens Medical Solutions USA Inc., Charlestown, MA, United States, ²A. A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown, MA, United States, ³Siemens Healthcare, Erlangen, Germany

In this work we developed a novel prospective motion correction method for multi-slice single shot diffusion weighted EPI. Rigid body navigation is achieved using non diffusion encoded low resolution single shot EPI images as motion navigators during the diffusion scan. Two approaches: integrated and interleaved are described. The proposed methods work independent of the b-value used and do not need retrospective adjustment of the b-matrix. The compromise for the integrated method is the TE and TR increase and the corresponding signal decrease while the interleaved method requires only a small (~10%) increase in minimum TR.

Murat Aksoy¹, Melvyn Ooi¹, Samantha J Holdsworth¹, Rafael O'Halloran¹, and Roland Bammer^1
1Center for Quantitative Neuroimaging, Department of Radiology, Stanford University, Stanford, CA, United States

Two prospective motion correction techniques for high resolution diffusion-weighted imaging are compared for readout-segmented (RS) EPI. The first uses 3D registration of low resolution navigator images and the second uses a camera mounted in the scanner bore to correct for motion between k-space segments. Results show that both techniques are effective in eliminating head motion between blinds of the RS-EPI readout, however, optical tracking has advantages because it needs less rescanning, is not vulnerable to intra-volume motion and optically obtained motion estimates are robust to ghosting artifacts.

Rita G. Nunes^1,2, Shaihan J. Malik², and Joseph V. Hajnal^2
1Institute of Biophysics and Biomedical Engineering, Faculty of Sciences, University of Lisbon, Lisbon, Portugal, ²Robert Steiner MRI Unit, Imaging Sciences Department, MRC Clinical Sciences Centre, Hammersmith Hospital, Imperial College London, London, United Kingdom

Unlike Echo-planar imaging, single-shot fast spin-echo is insensitive to field inhomogeneities but it requires precise control of the phase of the magnetization prior to the start of the refocusing train (CPMG condition). Due to motion, this is very difficult to achieve when diffusion-weighting is applied in vivo. Previously it has been shown that linear phase errors can be measured and corrected for in real time using gradients. The remaining non-linear phase modulations can be prospectively corrected using tailored RF excitation pulses as demonstrated on phantoms. Here the method is further developed and shown to be effective for in vivo imaging.

Robert Frost¹, David A. Porter², Gwenaelle Douaud¹, Peter Jezzard¹, and Karla L. Miller^1
1FMRIB Centre, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, United Kingdom, ²Healthcare Sector, Siemens AG, Erlangen, Germany

Readout-segmented EPI (rs-EPI) for diffusion imaging reduces distortion and T2* blurring and enables higher resolution relative to single-shot EPI (ss-EPI). However, the long scan times caused by segmenting the acquisition of k-space limit the number of slices and/or diffusion directions. A blipped-CAIPI multiband modification to rs-EPI has the potential to address the scan time issue and is demonstrated here with slice acceleration factor two. Trace-weighted images with 0.9x0.9x4mm resolution acquired in clinically relevant scan times are presented. Diffusion tractography is compared from data acquired at 2 and 1.5mm isotropic resolution for rs- and ss-EPI.

Mathias Engström¹, Roland Bammer², and Stefan Skare^1
1Clinical Neuroscience, Karolinska Institute, Stockholm, Sweden, ²Radiological Sciences Laboratory, Stanford University, Palo Alto, CA, United States

In this work we present a method to do diffusion weighted imaging at 1.3 mm isotropic resolution with full brain coverage, using multi-slab multi-echo spin echo-EPI. We also suggest a method on how to correct for diffusion gradient induced phase for multi-slab acquisitions along with slab profile intensity optimization.

Rafael O'Halloran¹, Murat Aksoy¹, and Roland Bammer^1
1Radiology, Stanford Universtiy, Stanford, CA, United States

The main obstacle to high-resolution 3D diffusion-weighted MRI is the motion-induced phase error. In this work the phase error is addressed with a hybrid 3D navigator approach that corrects phase induced by rigid-body motion for every shot and phase induced by repeatable non-rigid-body pulsation over the cardiac cycle. This phase correction method was shown to mitigate signal dropouts caused by shot-to-shot phase inconsistencies compared to a standard gridding reconstruction in healthy volunteers. The 3D SSFP approach was also compared to 2D DW-EPI and shown to have similar diffusion contrast.

Chu-Yu Lee^1,2, Zhiqiang Li³, James G. Pipe², and Josef P. Debbins^1,2
1Electrical Engineering, Arizona State University, Tempe, AZ, United States, ²Neuroimaging Research, Barrow Neurological Institute, Phoenix, AZ, United States,³MR engineering, GE Healthcare, Waukesha, WI, United States

Compared with conventional (multishot FSE) DW-PROPELLER, Turboprop gives increased sampling efficiency, a wider self-navigated region, and reduced specific absorption rate (SAR) by incorporating the GRASE readout to collect gradient echoes around the primary spin-echo. However, phase errors using the GRASE readout, which are exacerbated with preceding large diffusion gradients, induce image artifacts in Turboprop. To mitigate this issue, X-prop and Steer-prop techniques have been proposed, which keep the gradient echoes encoded into separate blades. In this work, we introduced a method to correct the off-resonance phase in Turboprop, called ¡¥Turboprop+¡¦. The results suggest that Turboprop+ has greater immunity to the artifacts from off-resonance phase, compared with X-prop.

Jedrzej Burakiewicz¹, Geoffrey David Charles-Edwards^1,2, Vicky Goh^1,2, and Tobias Schaeffter^1
1King's College London, London, United Kingdom, ²Guy's and St. Thomas' NHS Trust, London, United Kingdom

Effective fat suppression can significantly extend the acquisition time in EPI-based diffusion MRI outside the brain. We present a single-shot diffusion-weighted EPI in combination with IDEAL method for fat-water sepa-ration utilising the inherent signal averaging from the chemical shift encoding. We utilise a B0 map acquired without diffusion weighting to demodulate images at higher b-values before reconstruction to improve the wa-ter-fat separation. This technique was tested in phantom and healthy volunteers.

Steven Baete¹, Gene Cho^1,2, and Eric E Sigmund^1
1Center for Biomedical Imaging, Dept. of Radiology, NYU Langone Medical Center, New York, NY, United States, ²Sackler Institute of Graduate Biomedical Sciences, NYU School of Medicine, New York, NY, United States

This abstract describes the implementation of a rapid method to acquire a full diffusion tensor in on a clinical scanner. The method is named the 3D multiple modulation multiple echo diffusion tensor acquisition technique (MEDITATE). MEDITATE employs four rf-pulses and a pattern of diffusion gradients on three gradient axes to encode a train of 17 echoes with different diffusion weightings and directions. The resulting diffusion weighted signals can be used to estimate DTI parameters as demonstrated in a fibrous phantom. This sequence may be useful in clinical applications requiring time-sensitive acquisition of DTI parameters such as dynamical DTI in muscle.