Franciszek Hennel¹ and Klaas P. Pruessmann^1
1Institute for Biomedical Engineering, University of Zurich and ETH Zurich, Zurich, Switzerland

Classically encoded MRI signals are complex and therefore sensitive to uncontrolled phase variations. We propose an alternative spatial encoding method which leads to real positive signals and allows phase fluctuations to be removed by a simple magnitude calculation before the Fourier transform. The phase immunity of the method is demonstrated by recovering an image from a scan with unknown random receiver phase.

James C Korte¹, Bahman Tahayori¹, Peter M Farrell¹, Stephen M Moore^2,3, and Leigh A Johnston^1
1Dept. Electrical and Electronic Engineering, University of Melbourne, Melbourne, Australia, ²IBM Research, Melbourne, Australia, ³Dept. Mechanical Engineering, University of Melbourne, Melbourne, Australia

The observable periodic magnetisation induced in a spin system excited by Rabi modulated Continuous Wave excitation is exploited in this work to construct a new imaging paradigm.  Localised frequency information is encoded in the steady-state Rabi harmonics, reconstructed as radial projections of proton density and back-projected to form images.  This form of imaging has the potential to image samples with ultra-short T₂ decay, which is beneficial for the diagnosis of muscular skeletal injury and disease.

Somaie Salajeghe¹, Paul Babyn², Logi Vidarsson³, and Gordon E. Sarty^1
1Biomedical Engineering, University of Saskatchewan, Saskatoon, SK, Canada, ²Medical Imaging, University of Saskatchewan, Saskatoon, SK, Canada, ³LT Imaging, Toronto, ON, Canada

Portable MRI can be possible by eliminating gradient coils and B₀ homogeneity requirements. Relaxing the B₀ homogeneity requirements leads to non-uniform B₀ field. In-homogeneous B₀ fields have the potential to encode spatial information in one direction for use in novel image encoding schemes. We investigated the possibility of image reconstruction of the signal from a non-uniform rotating magnetic field and two rotating RF receivers. Our results indicate that this is a feasible approach. 

John Stuart Haberl Baxter¹, Zahra Hosseini¹, Junmin Liu², Maria Drangova³, and Terry M Peters^1
1Biomedical Engineering Graduate Program, Western University, London, ON, Canada, ²Imaging Laboratories, Robarts Research Institute, London, ON, Canada, ³Department of Medical Biophysics, Western University, London, ON, Canada

Tissue susceptibility differences manifest in MR phase images as high-frequency changes in an otherwise smooth phase background. Two paradigms currently exist for isolating these changes: one involves phase unwrapping followed by filtering; the other involves filtering the complex signal. Both rely on a linear topology, which can result in artifacts such as phase wraps and shadowing, as phase is inherently cyclic. This paper introduces the cyclic continuous max-flow (CCMF) method, which uses optimization over a cyclic topology to process phase information. More robust field maps are generated using this approach compared to the traditional paradigms.

Madison Kretzler¹, Jesse Hamilton², Mark Griswold^2,3, and Nicole Seiberlich^2,3
1Electrical Engineering, Case Western Reserve University, Cleveland, OH, United States, ²Biomedical Engineering, Case Western Reserve University, Cleveland, OH, United States, ³Radiology, University Hospitals, Cleveland, OH, United States

This abstract presents a-f BLAST, a non-iterative approach to non-Cartesian k-t BLAST for radial trajectories, and demonstrates its use for accelerated cardiac imaging.

Zhengguo Tan¹, Volkert Roeloffs¹, Dirk Voit¹, Arun Joseph¹, Markus Untenberger¹, Klaus-Dietmar Merboldt¹, and Jens Frahm^1
1Biomedizinische NMR Forschungs GmbH, Max-Planck-Institute for Biophysical Chemistry, Goettingen, Germany

The proposed model-based reconstruction technique jointly computes a magnitude image, a phase-contrast map, and a set of coil sensitivities from every pair of flow-compensated and flow-encoded datasets obtained by highly undersampled radial FLASH. Real-time acquisitions with 5 and 7 radial spokes per image resulted in 25.6 and 35.7 ms measuring time per phase-contrast map, respectively. It yields quantitatively accurate phase-contrast maps with improved spatial acuity, reduced phase noise, reduced partial volume effects, and reduced streaking artifacts.

Jan-Willem Beenakker¹, Lucia Hervella², Juan Tabarnero², Dennis Shamonin¹, Andrew Webb¹, Gregorius Luyten¹, and Pablo Artal^2
1Leiden University Medical Centre, Leiden, Netherlands, ²University of Murcia, Murcia, Spain

Patient-specific three-dimensional eye models obtained using very high resolution scans on a human 7T MRI system have been shown to form a much more accurate input for ray tracing algorithms than the current state-of-the-art generalized eye models used for clinical ophthalmology. Using a cued-blink protocol, custom-built phased array coil and segmentation software, accuracy of less than one-half dioptre can be achieved using the MRI data. These patient-specific models should provide much improved input for therapeutic procedures such as intra-ocular lens replacement for post-cataract surgery.

Yunjie Tong¹, Kimberly P Lindsey¹, Lia M Hocke², Gordana Vitaliano¹, Dionyssios Mintzopoulos¹, and Blaise B Frederick^1
1McLean Hospital/Harvard Medical School, Belmont, MA, United States, ²Hotchkiss Brain Institute, University of Calgary, Calgary, AB, Canada

Previously, we have demonstrated that we can extract systemic low frequency oscillation (sLFO) from resting state (RS) fMRI data and map its dynamic patterns as it moves through the brain. We have hypothesized that the dynamic patterns represent the cerebral blood flow. In this study, we tested this hypothesis by conducting both Dynamic Susceptibility Contrast scan (bolus tracking) and RS fMRI scan in health subjects. By comparing the flow patterns of the bolus with that of sLFO, we found that the flow of sLFO does represent the blood flow, however, mostly in the capillaries and veins.

Yuqing Wan¹, Maolin Qiu¹, Gigi Galiana¹, and R. Todd Constable^1
1Radiology and Biomedical Imaging, Yale University, New Haven, CT, United States

We developed a nonlinear encoding method with multiple RF coils based on the Bloch-Siegert shift. Simulated reconstructions showed that higher B1 fields and lower off-resonance frequency shift improves reconstruction quality. This approach is potentially promising as a replacement for conventional gradient encoding providing excellent spatial encoding with essentially silent imaging.

Kristy Tan¹, Adam L. Sandler², Avital Meiri¹, Rick Abbott², James T. Goodrich², Eric Barnhill³, and Mark E. Wagshul^1
1Gruss MRRC, Albert Einstein College of Medicine, Bronx, NY, United States, ²Department of Neurological Surgery, Albert Einstein College of Medicine/Children’s Hospital at Montefiore, Bronx NY, Bronx, NY, United States, ³Clinical Research Imaging Centre, University of Edinburgh, Edinburgh, United Kingdom

Hydrocephalus patients with functioning shunts are often faced with severe headache disorders. This is believed to be due to a change in brain viscoelasticity. MRE uses external mechanical vibrations to induce waves and estimates viscoelasticity from the wave propagation. This study found a significant decrease of brain viscoelasticity in patients (N=14) compared to controls (N=12) (G* white matter, controls: 1407.82 (SD=111.3) Pa vs patients: 1099.33 (SD=262.86) Pa, p =0.0001). Additionally, an inverse correlation between ventricular volume and viscoelasticity in corresponding lobes was found indicating that brain viscoelasticity may play a role in hydrocephalus patient’s symptoms such as headaches.

Alexander Aranovitch¹, Maximilian Haeberlin¹, Simon Gross¹, Thomas Schmid¹, and Klaas Paul Pruessmann^1
1Institute for Biomedical Engineering, ETH Zurich and University of Zurich, Zurich, Switzerland

A field-detection based method for prospective motion correction is proposed which uses the sequence itself for localizing NMR field probes. No additional gradients or increase of the sequence duration are required to apply this method to various MR sequences, such as clinically relevant spin-warp sequences. The proposed method collects high-frequency information present due to gradient switching from multiple short, temporally separated snippets within one TR. A precision on the order of 10µm and 0.01° (RMS) for translational and rotational degrees of freedom is obtained. The method is demonstrated in-vivo with high-resolution T2*-weighted gradient echo scans. 

Justin P Haldar¹, Qiuyun Fan², and Kawin Setsompop^2
1Electrical Engineering, University of Southern California, Los Angeles, CA, United States, ²A. A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown, MA, United States

We propose a novel approach to data acquisition and image reconstruction that achieves high-quality in vivo whole-brain human diffusion imaging at (660 µm)³ resolution in 25 minutes.  The approach uses a powerful acquisition strategy (generalized SLIce Dithered Enhanced Resolution Simultaneous MultiSlice, or gSlider-SMS) that enables high-resolution whole-brain imaging in 25 minutes (64 diffusion weightings + 7 b=0 images), but the resulting images suffer from low SNR without averaging.  To address the SNR problem, we utilize a regularized reconstruction/denoising approach that leverages the shared spatial structure of different diffusion images.  In vivo results demonstrate the effectiveness of this approach.

Stephen Cauley^1,2, Kawin Setsompop^1,2, Berkin Bilgic¹, Himanshu Bhat³, Borjan Gagoski^2,4, Thomas Witzel^1,2, and Lawrence L. Wald^1,2,5
1MGH/HST, Athinoula A. Martinos Center for Biomedical Imaging, Charlestown, MA, United States, ²Harvard Medical School, Boston, MA, United States, ³Siemens Medical Solutions Inc, Malvern, PA, United States, ⁴Fetal-Neonatal Neuroimaging & Developmental Science Center, Boston Children's Hospital, Boston, MA, United States, ⁵Harvard-MIT Division of Health Sciences and Technology, MIT, Cambridge, MA, United States

Fast MRI acquisitions often rely on efficient traversal of k-space, e.g. Spiral, EPI, and Wave-CAIPI. Limitations in hardware and other physical effects cause these trajectories to deviate from the theoretical path, and additional measurements are typically used to approximate discrepancies. We propose a joint optimization to directly estimate trajectory discrepancies simultaneously with the underlying image, without need for additional characterization measurements. Model reduction schemes are introduced to make this optimization computationally efficient and ensure final image quality. We demonstrate our approach for a clinically relevant Wave-CAIPI acquisition, where we accurately optimize across >6million unknowns in 30s on standard vendor hardware.

Ana Beatriz Solana¹, Anne Menini¹, and Florian Wiesinger^1
1GE Global Research, Garching bei Muenchen, Germany

Zero TE is an extremely efficient 3D pulse sequence which also has the advantages of low geometrical distortion, reduced acoustic noise and the capacity of imaging short T2 structure. However, its native contrast is proton density. Here we present a novel method that allows gradient refocusing at echo times suitable for fMRI or susceptibility weighted imaging. As a proof of concept, this new imaging strategy is tested in phantom experiments.

Muhammad Faisal Siddiqui¹, Abubakr Shafique², Yousif Rauf Javed², Talha Ahmad Khan², Hamza Naeem Mughal², Ahmed Wasif Reza¹, Hammad Omer², and Jeevan Kanesan^1
1Electrical Engineering, University of Malaya, Kuala Lumpur, Malaysia, ²Electrical Engineering, COMSATS Institute of Information Technology, Islamabad, Pakistan

FPGA (Field Programmable Gate Array) based application specific hardware, for real-time Sensitivity Encoding (SENSE) reconstruction, embedded on the receiver coil system may provide reconstruction without transferring the data to the MRI server. This may dramatically decrease the transmission cost of the system and the image reconstruction time. This paper proposes an FPGA implementation of SENSE algorithm using two different sensitivity maps estimation methods (pre-scan and E-maps). The results show that the proposed system consumes  only 145.64 μs for SENSE reconstruction (acceleration factor=2), while maintaining the quality of the reconstructed images with good mean SNR (29+ dB) and significantly less artefact power (<9×10^-4) values.

Benedikt A Poser^1
1Faculty of Psychology and Neuroscience, Maastricht University, Maastricht, Netherlands

Parallel imaging with controlled aliasing has revolutionised the way we do MRI, and this may directly translate to the way we bake. In this work CAIPIRINHA principles are successfully applied to the baking of cinnamon rolls. Furthermore, the question is considered of whether CAIPIRINHA may have been inspired by established baking practices in the first place.