Ashley G Anderson III¹, Dinghui Wang¹, and James G Pipe^1
1Imaging Research, Barrow Neurological Institute, Phoenix, AZ, United States

Autofocusing was used to estimate residual off-resonance and remove associated blurring in spiral images after initial joint deblurring and water-fat separation with a previously acquired field map.

Mark BYDDER¹, Wafaa Zaaraoui¹, and Jean-Philippe Ranjeva^1
1Aix Marseille Université, MARSEILLE, France

Non-Cartesian reconstructions typically perform a convolution to interpolate irregularly spaced samples onto a regular grid. The number of coefficients in the convolution trades-off accuracy with speed. A radially symmetric convolution kernel provides a favorable trade-off as compared to a separable kernel with the same number of coefficients.

Lin Chen¹, Shuhui Cai¹, and Congbo Cai^1
1Department of Electronic Science, Xiamen Unviersity, Xiamen, China, People's Republic of

As an emerging ultrafast imaging method, spatiotemporally encoded single-shot MRI is advantageous because it can resist off-resonance effects while retaining spatial and temporal resolutions comparable to the classical EPI. In this work, a variable density spiral sampling (VDSS) scheme is proposed for SPEN MRI. An optimization algorithm is used to design the sinusoidal readout gradient waveform. A specific gridding algorithm and non-Cartesian super-resovled reconstruction are proposed to retrieve image. Compared to the Cartesian sampling, VDSS can provide images with less artifacts and better spatial resolution.

JeongTaek Lee^1,2, Jinil Park^1,2, and Jang-Yeon Park^1,2
1Center for Neuroscience Imaging Research, Institute for Basic Science, Suwon, Korea, Republic of, ²Department of Biomedical Engineering, Sungkyunkwan University, Suwon, Korea, Republic of

Radial-acquisition imaging is recently gaining more popularity. For RA image reconstruction, both fast Fourier transform via gridding and filtered back-projection are available. Because of the processing time, FFT is dominantly used. If the processing-time issue is resolved, 3D FBP can be promising in the case of FID sampling or echo sampling with mis-centering of k-space. In this study, we propose a strategy that can significantly reduce the reconstruction time of 3D FBP in combination with a new RA scheme. Performance of the proposed method was demonstrated in phantom and human brain imaging at 3T.

mahwish khan¹, Taquwa Aslam¹, and Hammad Omer^1
1Electrical Engineering, COMSATS institute of information technology, Islamabad, Pakistan

The estimation of receiver coil sensitivity profiles is required for many PMRI  algorithms including CG-SENSE. Conventionally,CG-SENSE uses pre-scan method to estimate the sensitivity maps for which a separate scan is required. The novelty in this work is the use of GROG gridding on the central region of the acquired under-sampled cardiac radial data to estimate the receiver coil sensitivities using Eigen-value method. The results show that  GROG based sensitivity map estimation (proposed method) is an effective method without any requirement of a prior scan or body coil image.

David S Smith¹, Saikat Sengupta¹, Aliya Gifford¹, and E Brian Welch^1
1Institute of Imaging Science, Vanderbilt University, Nashville, TN, United States

We achieve significantly improved image quality in whole-body continuous moving table MRI without time penalty by using a multichannel, golden angle radial acquisition coupled to a calibrationless reconstruction.

Charles G Cantrell¹, Parmede Vakil^1,2, Donald R Cantrell³, Yong Jeong¹, Sameer A Ansari³, and Timothy J Carroll^1,3
1Biomedical Engineering, Northwestern, Chicago, IL, United States, ²College of Medicine, University of Illinois, Chicago, IL, United States, ³Radiology, Northwestern, Chicago, IL, United States

We have found that performing iterative PLCG dramatically (on average 55% better) improves reconstruction quality of an MR-PARSE acquisition.  Moreover, iterative PLCG has shown to be capable of reconstruction in regions with large susceptibility artifact.   Consistent frequency measurements allow us to remove static offsets caused by air interfaces near the earholes and sinuses leaving dynamic frequency offsets which transpose linearly to OEF.  Furthermore, our approach to prevent local minima solutions, through the use of an iterative PLCG, represents a new approach that may improve upon many other complex reconstruction methods.

Ina Nora Kompan¹ and Matthias Guenther^2
1mediri GmbH, Heidelberg, Germany, ²Fraunhofer Mevis, Bremen, Germany

3D radial acquisitions are prone to gradient delay and phase artifacts, which need to be corrected for. Ideally, the phase can be retrospectively corrected for by assuming the same phase in k-space center where all projections meet. However, if data are affected by gradient delays, profiles are shifted at varying angles and might not intersect at all. Here, a simple to apply, novel phase correction method is presented in which the changed trajectory due to gradient delays is incorporated into 3D phase correction. It was shown that artifacts are reduced by 12% compared to only gradient delay corrections.

Di Cui¹, Hing-Chiu Chang¹, Hua Guo², Queenie Chan³, and Edward S Hui^1
1Diagnostic Radiology, The University of Hong Kong, Hong Kong, Hong Kong, ²Department of Biomedical Engineering, Tsinghua University, Beijing, China, People's Republic of, ³Philips Healthcare, Hong Kong, Hong Kong

Magnetic Resonance Fingerprinting (MRF) enables simultaneous quantification of multiple relaxation parameters, which is further accelerated by usage of simultaneous multi-slice (SMS) technique. SENSE based SMS-MRF reconstruction suffers high g-factor penalty due to similar coil sensitivity profiles of collapsed slices. Here we proposed a new method to solve this problem, by data re-grouping from adjacent time points, spatial controlled aliasing is enabled, after dictionary matching, good parameter mapping of 2 slices with nearly the same sensitivity is acquired at half the acquisition time of single-excitation MRF.

Dinghui Wang¹, Zhiqiang Li¹, Ryan K. Robison¹, and James G. Pipe^1
1Imaging research, Barrow Neurological Institute, Phoenix, AZ, United States

Spiral in-out readout is an efficient sampling scheme for T2-weighted (T2w) spin-echo (SE) sequences. Two sets of spiral in-out data at different TEs are typically acquired so that deblurred water and fat images can be extracted from the spiral-in and the spiral-out data separately, which are then combined together. A method has been recently proposed to reconstruct water and fat images from a single set of spiral in-out data. This work demonstrates the feasibility of using this method as a fast and scan efficient means of fat suppression to reduce the scan time of the spiral T2w SE by a factor of 2.

Maria Engel¹, Lars Kasper¹, Christoph Barmet¹, Klaas Paul Prüssmann¹, and Thomas Schmid^1
1Institute for Biomedical Engineering, University Zurich and ETH Zurich, Zurich, Switzerland

Fast spiral sequences with concurrent field monitoring, B0 correction and SENSE-reconstruction promise a brave new world for time series MRI.

Melvyn B Ooi^1,2, Dinghui Wang², Ashley G Anderson III², Zhiqiang Li², Nicholas R Zwart², Ryan K Robison², and James G Pipe^2
1Philips Healthcare, Cleveland, OH, United States, ²Imaging Research, Barrow Neurological Institute, Phoenix, AZ, United States

Spiral MRI enables long sampling durations but at the cost of increased sensitivity to B0-field fluctuations. For example, scanner B0-drift will be observed in spiral MRI as a time-varying component of image blurring. A spiral deblurring strategy is presented where a reference B0 map is acquired at the start of the spiral exam; F0 navigators then quickly measure B0 drift over the course of the spiral exam, and are used to calculate B0-drift corrected B0-maps for deblurring of the current scan. F0-navigator accuracy is verified with independent B0-map acquisitions, and improved spiral deblurring is shown for a structural brain scan.

Ibtisam Aslam¹, Faisal Najeeb¹, and Hammad Omer^1
1Electrical Engineering, COMSATS Institute of Information Technology, Islamabad, Pakistan

Accelerated non-Cartesian parallel imaging plays a vital role to reduce data acquisition time in the MR imaging; however the resultant images may contain aliasing artifacts. In this work, the application of ESPIRiT with GROG is proposed to get good reconstruction results from highly under-sampled radial data. The proposed method is tested on 3T short–axis cardiac radial data at different acceleration factors (AF=4, 6 and 9) and compared with pseudo-Cartesian GRAPPA. The results show that the proposed method offers significant improved (e.g. 81% improvement in term of artifact power at AF=4) reconstruction results as compared to conventional pseudo-Cartesian GRAPPA.

Zhifeng Chen¹, Liyi Kang¹, Allan Jin², Feng Liu³, Ling Xia¹, and Feng Huang^2
1Biomedical Engineering, Zhejiang University, Hangzhou, China, People's Republic of, ²Philips Healthcare (Suzhou) Co. Ltd, Suzhou, China, People's Republic of, ³School of Information Technology and Electrical Engineering, The University of Queensland, Queensland, Australia

Dynamic contrast enhanced (DCE) MRI plays an important role in the detection of liver metastases, characterization of tumors, assessing tumor response and studying diffuse liver disease. It requires a high spatial-temporal resolution. Existing iterative dynamic MRI reconstruction algorithms, such as iGRASP and L+S, realize their functions through iterative schemes. Though the solutions are generally acceptable, yet suffer from significantly high computational cost. This study proposed to use dynamic artificial sparsity and non-Cartesian parallel imaging for high spatiotemporal resolution DCE reconstruction, which results in comparable image quality relative to the above iterative schemes with greatly reduced computational cost.

Taquwa Aslam¹, Mahwish Khan¹, Ali Raza Shahid¹, and Hammad Omer^1
1Electrical Engineering, COMSATS Institute of information technology, Islamabad, Pakistan

CG SENSE is an iterative algorithm used in PMRI for MR image reconstruction from under-sampled data. One major limitation of CG-SENSE is the appropriate choice of the number of iterations required for good reconstruction results.This paper proposes the use of a correlation measure between the line profiles of the reconstructed images in the current and the previous iteration, as a stopping criterion for the CG-SENSE algorithm. Results of the proposed method are compared with the Bregman distance stopping criterion. The results show that the use of line profile correlation measure acts as an effective stopping criterion in CG-SENSE.

Jyh-Miin Lin¹, Andrew J. Patterson², Chung-Wei Lee³, Ya-Fang Chen³, Tilak Das⁴, Daniel Scoffings⁵, Hsiao-Wen Chung⁶, Jonathan H. Gillard¹, and Martin J. Graves^2
1Department of Radiology, University of Cambridge, Cambridge, United Kingdom, ²MRIS unit, Cambridge University Hospitals NHS Foundation Trust, Cambridge, United Kingdom, ³Department of Radiology, National Taiwan University College of Medicine, Taipei, Taiwan, ⁴Addenbrooke’s Hospital, Cambridge, United Kingdom, ⁵Department of Radiology, Cambridge University Hospitals NHS Foundation Trust, Cambridge, United Kingdom, ⁶Department of Electrical Engineering, National Taiwan University, Taipei, Taiwan

PROPELLER MRI can reduce motion artifacts. However, the colored noise of PROPELLER could degrade the image quality. Although the iterative Pseudo-Inverse with COnstraints (PICO) has been proposed to improve the image quality metrics, further clinical validation is needed. In this study, two neuroradiologists compared the image quality of PICO with the standard density compensation. Results show that PICO significantly improves the identification of two anatomical structures and the clinical utility. 

Kousaku Saotome¹, Akira Matsushita¹, Koji Matsumoto², Yoshiaki Kato³, Kei Nakai⁴, Yoshiyuki Sankai⁵, and Akira Matsumura^4
1Center for Cybernics Research, University of Tsukuba, Tsukuba, Japan, ²Department of Radiology, Chiba University Hospital, Chiba, Japan, ³Medical Technology Department, Kameda General Hospital, Kamogawa, Japan, ⁴Department of Neurosurgery, University of Tsukuba, Tsukuba, Japan, ⁵Faculty of Engineering, Information and Systems, University of Tsukuba, Tsukuba, Japan

Our purpose is to develop the novel brain phantom including low contrast and to verify its potential to emphasize the motion correction effects in PROPELLER. Our proposed phantom set would allow not only to add the ability to low contrast objects but also to provide exact rotations instead of a healthy volunteer. The making process of the phantom consists of making profile curves, transforming to depths of the convexo-concave, printing using 3-D printer, and pouring agarose. This is the novel phantom making process.

Rolf Pohmann¹ and Klaus Scheffler^1,2
1Magnetic Resonance Center, Max Planck Institute for Biological Cybernetics, Tübingen, Germany, ²Biomedical Magnetic Resonance, University Tübingen, Tübingen, Germany

Averaging is a frequently used way to increase the SNR of a measurement. Here we show that spending the additional time for increasing the spatial resolution and applying a retrospective k-space filter can yield a higher SNR gain than conventional averaging. For weighting in two phase encoding directions, this can increase the SNR by up to 25%. For 3D weighting, SNR gain can reach 57%, if the additional acquired k-space points are used to increase the readout duration, and 38% for equal duration for weighted and unweighted acquisition.

Rong Guo¹, Yicheng Chen², Jinsong Ouyang³, Georges El Fakhri³, and Kui Ying^1
1Engineering Physics, Tsinghua University, Beijing, China, People's Republic of, ²University of California, Berkeley, Berkeley, CA, United States, ³Center for Advanced Radiological Sciences, Massachusetts General Hospital, Boston, MA, United States

Joint reconstruction of PET and MRI is aimed to improve both PET and MR image quality using information from each other’s imaging modality. However, the aim cannot be achieved without correcting the attenuation of PET. We proposed a method to correct attenuation during the process of simultaneous PET and MRI images reconstruction. This method integrates PET, MRI and TOF information.

Seul Lee¹ and Gary Glover^2
1Department of Electrical Engineering, Stanford University, Stanford, CA, United States, ²Department of Radiology, Stanford University, Stanford, CA, United States

Since half k-space reconstruction reduces scan time while keeping spatial resolution, it can be used for T₂* weighted images such as functional MRI that requires fairly long TE. Functional MRI is sensitive to off-resonance because there exist large susceptibility variations in air-tissue interfaces such as sinuses. Existing half k-space reconstruction is vulnerable to off-resonance since it might lose most of the image energy when there is a large amount of phase shift. In this study, we suggest a new half k-space reconstruction method that is more robust to off-resonance compared to existing reconstruction method.

Josip Marjanovic¹, Jonas Reber¹, David Otto Brunner¹, Bertram Jakob Wilm^1,2, and Klaas Paul Pruessmann^1
1Institute for Biomedical Engineering, University and ETH Zurich, Zurich, Switzerland, ²Skope Magnetic Resonance Technologies, Zurich, Switzerland

Data amounts of massive parallel receiver arrays as well as latency requirements of real-time applications such as interventional MRI, navigators etc. prompt for high-speed data preprocessing. First steps in the reconstruction such as noise pre-whitening or channel combinations and compressions can be efficiently performed on FPGAs. Here we present a flexible system and software architecture for such tasks and demonstrate its capability performing real-time coil compression directly in the spectrometer.

Yong Liu¹ and Paul R. Harvey^2
1Philips Research China, Shanghai, China, People's Republic of, ²Philips MR, Best, Netherlands

Extraneous in-band Radio Frequency Interference (RFI) signals can mix with the MR signal of a subject creating artefacts in the image data, usually zipper like artefacts. In order to eliminate/reduce the impact of RFI, this abstract describes a software de-noising method which can be implemented both online and off-line through k-space manipulation.

Jinil Park^1,2, Taehoon Shin³, Soon Ho Yoon^4,5, Jin Mo Goo^4,5,6, and Jang-Yeon Park^1,2
1Center for Neuroscience Imaging Research, Institute for Basic Science, Suwon, Korea, Republic of, ²Department of Biomedical Engineering, Sungkyunkwan University, Suwon, Korea, Republic of, ³Diagnostic Radiology and Nuclear Medicine, University of Maryland, Baltimore, MD, United States, ⁴Department of Radiology, Seoul National University College of Medicine, Seoul, Korea, Republic of, ⁵Institute of Radiation Medicine, Seoul National University Medical Research Center, Seoul, Korea, Republic of, ⁶Cancer Research Institute, Seoul National University, Seoul, Korea, Republic of

Radial-acquisition imaging is suitable for lung imaging because it allows a very short TE and a desirable degree of motion insensitivity. Despite its tolerance to motion artifacts, respiratory motion is still a major reason of causing image artifacts in lung imaging, which motivates the need for the respiratory gating. In retrospective-respiratory gating, it is significant to keep the k-space as uniform as possible after gating. In this study, we explore the optimal condition for obtaining uniform k-space coverage after retrospective gating in consideration of various breathing patterns.

Lijun Bao^1
1Department of Electronic Science, Xiamen University, Xiamen, China, People's Republic of

In the learning based single image super-resolution restoration, the high frequency information is enhanced by retrieving the high-frequency information from the high resolution training samples. Therefore how to reveal the underlying relations between the HR and the LR patch spaces is the key issue. In this work, we propose to cluster the pre-collected HR example patches to generate subdictionary and select the proper subdictionary for any image patch according to the frequency spectrum feature in Fourier domain, because the Fourier spectrogram can reflect the feature complexity, local directionality and the texture periodicity of the image patch simultaneously.

Weiran Deng¹, Michael Herbst¹, and V. Andrew Stenger^1
1University of Hawaii JABSOM, Honolulu, HI, United States

A subset of fMRI images is used to reconstruct a structural image at a higher resolution using a Super-Resolution (SR) reconstruction method. The subset of fMRI images are selected such that the translation and rotation between the shots are less than the pixel size and therefore useful for modeling the pixel characteristics. The preliminary results demonstrate the feasibility of reconstructing a structural image with susceptibility contrast from a subset of fMRI images. 

Moosa Zaidi¹, Joseph Cheng¹, Tao Zhang¹, and John Pauly^1
1Electrical Engineering, Stanford University, Stanford, CA, United States

Sensitivity to motion remains a major limitation to the clinical utility of MRI. Self-navigating Cartesian trajectory (Butterfly) can provide coil-by-coil estimates of local linear translation without lengthening scan time or requiring external sensors.  We propose to combine translational motion estimates with the geometry of the differing coil sensitivities to estimate global translation and rotation.  These estimates can then be used to retrospectively correct for motion.  For 2D slices we are successfully able to extract both unknown rotation with known translation and unknown translation and unknown rotation.  Extension to 3D is a promising direction for future work.

Viktor Vegh¹, Kieran O'Brien², David C Reutens¹, Steffen Bollmann¹, and Markus Barth^1
1Centre for Advanced Imaging, University of Queensland, Brisbane, Australia, ²Magnetic Resonance, Siemens Healthcare Pty. Ltd., Brisbane, Australia

Signal phase acquired via gradient recalled echo MRI sequences provides and an important source of tissue contrast. The use of phased array coils results in multiple-channel images that have to be combined to form a single image. A robust method of computing phase images has been challenging to develop, primarily due to the distribution of noise in phase images. We propose a new approach of combining phase images by exploiting the inherent noise in signal phase. Our selectively combined signal phase results show an improvement in the quality of the combined phase image in comparison to existing methods.

David S Smith¹, Saikat Sengupta¹, Aliya Gifford¹, and E Brian Welch^1
1Institute of Imaging Science, Vanderbilt University, Nashville, TN, United States

We present a software system called CITRON that can extract, reconstruct, and display images from non-Cartesian MR data at 60 frames per second, allowing the user to vary the reconstruction parameters in real time in order to explore and optimize the reconstructed images interactively.

Christine Law¹, Gary Glover¹, and Sean Mackey^1
1Stanford University, Stanford, CA, United States

We present a novel idea for quickly detecting the optimal half k-space for use in partial Fourier acquisition.  With EPI acquisition, the center of k-space can be offset from the origin by local magnetic field inhomogeneity.  This offset can occur in both positive and negative phase encode directions.  For partial Fourier acquisition, it is important to sample the portion of k-space containing the center peak.  Before data collection using partial Fourier acquisition, a reference scan that collects two time frames (each with different halves of k-space coverage) can be used to determine the proper half of k-space to collect for each slice.

Nicholas R. Zwart¹, Ashley G. Anderson III¹, Ryan K. Robison¹, Andrew Li², Mariya Doneva^2,3, Frank Ong², Martin Uecker², Michael Lustig², and James G. Pipe^1
1Imaging Research, Barrow Neurological Institute, Phoenix, AZ, United States, ²Electrical Engineering, University of California, Berkeley, CA, United States, ³Philips Research, Hamburg, Germany

This work proposes the use of the software development platform called GPI (Graphical Programming Interface) as a tool for resourcing other work for integration and comparison.  The GPI software structure is designed to facilitate the encapsulation of outside libraries and provides a plug-in model to isolate package dependencies.  The library featured in this work is the Berkley Advanced Reconstruction Toolkit (BART) which provides, multi-platform compatible, compressed sensing and parallel imaging algorithms.

Anne Slawig¹, Tobias Wech¹, Johannes Tran-Gia^1,2, Henning Neubauer¹, Thorsten Bley¹, and Herbert Köstler^1
1Departement for Diagnostic and Interventional Radiology, University of Würzburg, Würzburg, Germany, ²Department of Nuclear Medicine, Würzburg, Germany

Imaging of the beating heart is one of the main challenges in fast MRI. Balanced steady state free precession sequences are fast, yield high signal and have excellent contrast between blood and myocardium. Unfortunately, they are sensitive to field inhomogeneities, which lead to banding artifacts, considerably reducing the image quality. As the steady state tolerates small shifts in frequency it is possible to acquire a frequency- modulated bSSFP. Our study uses such a frequency-modulated approach in combination with a multi-frequency reconstruction to obtain banding free CINE images, with high signal and good contrast.

Stephan A.R. Kannengiesser¹, Boris Mailhe², Mariappan Nadar², Steffen Huber³, and Berthold Kiefer^1
1MR Application Predevelopment, Siemens Healthcare, Erlangen, Germany, ²Medical Imaging Technologies, Siemens Healthcare, Princeton, NJ, United States, ³Radiology and Biomedical Imaging, Yale University School of Medicine, New Haven, CT, United States

Spatially varying noise limits acquisition speed and spatial resolution in multi-channel MRI. Conventional single-slice noise-suppressing image filters without additional knowledge about data acquisition, image reconstruction, and noise level, have limited performance and need parameter tuning. In this work, an iterative denoising algorithm is presented which works with standard settings on 3D complex-valued data with supplementary information from the scanner environment.

Initial results from routine clinical imaging are promising: spatially adaptive, as intended, and superior to a commercially available image filter. Non-optimized reconstruction times of up to 15min per volume still need improvement, and further clinical investigations will be performed.

Uten Yarach¹, Hendrik Mattern¹, Alessandro Sciarra¹, and Oliver Speck^1
1Department of Biomedical Magnetic Resonance, Otto-von-Guericke University Magdeburg, Magdeburg, Germany

Phase image reconstruction from multi-channel data at high field strength becomes challenging since a volume body coil that provides the phase offset information is generally not available. Several coil combination techniques are still computation intensive and prone to errors for strongly T2*-weighted data due to the requirement of spatial smoothness of image phases, required phase reference from body coil scan, and also hampered for strongly localized B1 coil sensitivities. These limitations lead to phase image inconsistencies and may cause artifacts in QSMs. We demonstrated here that the coil combination should be considered after performing individual physical and/or virtual coil QSM.

Tetsuro Sekine^1,2, Ninon Burgos³, Geoffrey Warnock¹, Martin Huellner¹, Alfred Buck¹, Edwin ter Voert¹, M. Jorge Cardoso³, Brian Hutton³, Sebastien Ourselin³, Patrick Veit-Haibach¹, and Gaspar Delso^4
1University Hospital Zurich, Zurich, Switzerland, ²Nippon Medical School, Tokyo, Japan, ³University College London, London, United Kingdom, ⁴GE Healthcare, Waukesha, WI, United States

Accurate attenuation correction on PET/MR scanner is challenging. We compared multi-atlas method with clinical single-atlas method. Our study revealed that the error of PET images based on multi-atlas method is reduced from 1.5% to 1.2% compared to the single-atlas method, a 30% improvement.

 

Shanshan Wang¹, Zhenghang Su^1,2, Leslie Ying³, Xi Peng¹, and Dong Liang^1
1Shenzhen Institutes of Advanced Technologies, Shenzhen, China, People's Republic of, ²School of Information Technologies, Guangdong University of Technology, Guangzhou, China, People's Republic of, ³Department of Biomedical Engineering and Department of Electrical Engineering, The State University of New York, Buffalo, NY, United States

This paper proposes a deep learning based approach for accelerating MR imaging. With the utilization of a large number of existing high-quality MR images, we train an off-line convolutional neural network (CNN) to identify the mapping relationship between MR images obtained from zero-filled and fully-sampled k-space data. Then the trained CNN is employed to predict an image from undersampled data, which is used as the reference in solving an online constrained imaging problem. Results on in vivo datasets show that the proposed approach is capable of restoring fine details and presents great potential for efficient and effective MR imaging.

Congyu Liao¹, Ying Chen¹, Hongjian He¹, Song Chen¹, Hui Liu², Qiuping Ding¹, and Jianhui Zhong^1
1Department of Biomedical Engineering, Center for Brain Imaging Science and Technology, Zhejiang University, Hangzhou, China, People's Republic of, ²MR Collaboration NE Asia, Siemens Healthcare, Shanghai, China, People's Republic of

In this study, a rank constraint based Nyquist ghost removal method is proposed for single-shot spatiotemporally encoded (SPEN) MRI.

Sangtae Ahn¹, Lishui Cheng¹, Dattesh Shanbhag², and Florian Wiesinger^3
1GE Global Research, Niskayuna, NY, United States, ²GE Global Research, Bangalore, India, ³GE Global Research, Munich, Germany

PET attenuation correction is critical to accurate PET quantitation. For hybrid PET/MR imaging, MR-based attenuation correction (MRAC) has challenges in implants, internal air, bones and lung regions where MR signals are low. To address the challenges and improve robustness and accuracy of MRAC, a joint estimation algorithm with MR-based priors is implemented where prior weights are spatially modulated, providing great flexibility to users. The JE algorithm was applied to whole-body clinical TOF PET/MR data and it was demonstrated that the algorithm can recover the attenuation of implants, abdominal air and lungs in a robust way.

Thorsten Heußer¹, Christopher M Rank¹, Martin T Freitag², Heinz-Peter Schlemmer², Antonia Dimitrakopoulou-Strauss³, Thomas Beyer⁴, and Marc Kachelrieß^1
1Medical Physics in Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany, ²Department of Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany, ³Clinical Cooperation Unit Nuclear Medicine, German Cancer Research Center (DKFZ), Heidelberg, Germany, ⁴Center for Medical Physics and Biomedical Engineering, Medical University Vienna, Vienna, Austria

To improve attenuation correction (AC) and thus PET quantification for PET/MR imaging, we have recently proposed a method to jointly estimate attenuation and activity distributions from the non-TOF PET emission data. Available MR information is used to derive voxel-specific expectations on the attenuation coefficients, favoring the occurrence of pre-selected attenuation values corresponding to air, soft tissue, and bone. We here present first results for clinical non-TOF ¹⁸F-FDG PET/MR data sets of the head region. PET reconstruction was performed using MR-based AC as provided by the vendor, our proposed algorithm, and CT-based AC for comparison.

Jessica A McKay¹, Steen Moeller², Sudhir Ramanna², Edward J Auerbach², Michael T Nelson², Kamil Ugurbil², Essa Yacoub², and Patrick J Bolan^2
1Department of Biomedical Engineering, University of Minnesota, Minneapolis, MN, United States, ²Department of Radiology, University of Minnesota, Minneapolis, MN, United States

It is important to improve ghost correction in breast DWI using SE-EPI because residual ghosts bias ADC measurements, which reflect malignancy. The standard ghost correction method using a 3-line navigator frequently fails in breast DWI. In this work we implemented several alternative ghost correction strategies. The two best methods outperformed the standard 3-line navigator correction. These techniques applied 1D, linear corrections based on 1) phase maps from separate reference scans and 2) entropy minimization.  

Rong Guo¹, Pei Han¹, Yicheng Chen², Jinsong Ouyang³, Georges El Fakhri³, and Kui Ying^1
1Engineering Physics, Tsinghua University, Beijing, China, People's Republic of, ²UC Berkeley-UCSF Graduate Program in Bioengineering, University of California, Berkeley, Berkeley, CA, United States, ³Center for Advanced Radiological Sciences, Massachusetts General Hospital, Boston, MA, United States

The maximum likelihood activity and attenuation (MLAA) method usually utilizes time-of-flight (TOF) information to solve the problem of attenuation correction. However, the application of TOF brings noise. In this work, we proposed a method, Maximum a Posteriori for simultaneous activity and attenuation reconstruction (MAPAA), which introduces MRI information as prior knowledge into MLAA to reduce noise. 

Charles Guan¹ and Mark Chiew^2
1Electrical Engineering, Stanford University, Fremont, CA, United States, ²FMRIB Centre, University of Oxford, Oxford, United Kingdom

Functional MRI has been slow to benefit from data acceleration techniques based on non-linear image reconstruction. We present a comparison of two non-linear image reconstruction methods based on sparsity and low-rank models of FMRI data. k-t FOCUSS uses an asymptotic L1 minimization program to solve for a sparse x-f reconstruction. In contrast, k-t FASTER solves for a spatio-temporally low-rank reconstruction using an iterative hard thresholding and matrix shrinkage algorithm, without requiring a pre-specified basis. We applied each algorithm to incoherently sampled FMRI data and demonstrate that the strict rank-constraint method outperforms spectral- and Karhunen-Loeve Transform (KLT)-sparsity across different metrics.

Holger Eggers¹, Christian Stehning¹, Mariya Doneva¹, Elwin de Weerdt², and Peter Börnert^1,3
1Philips Research, Hamburg, Germany, ²Philips Healthcare, Best, Netherlands, ³Department of Radiology, Leiden University Medical Center, Leiden, Netherlands

3D Dixon TSE scans essentially provide the same information as several conventional 2D TSE scans in different orientation, without and with fat suppression. However, their scan time is usually still too long for clinical practice. In this work, the basic feasibility of accelerating a 3D Dixon TSE scan with PD weighting by a combination of compressed sensing and parallel imaging was investigated in knee imaging. Results obtained in half the scan time compared to the use of parallel imaging alone are presented, which indicate that 3D Dixon TSE scans may become as fast as current, conventional 3D TSE scans with fat suppression.

Luonan Wang¹ and Daniel S Weller^1
1Electrical and Computer Engineering, University of Virginia, Charlottesville, VA, United States

This abstract provides a joint motion estimation and image reconstruction method for data acquired using a spiral pulse sequence. It forms a data fitting term with image and motion variables and uses alternating minimization with conjugate gradients to solve the nonlinear optimization problem. This approach will allow MR scanning to be more robust to non-rigid motion while still achieve fast image reconstruction. The new approach will enable MR imaging with children without sedation.

Kinam Kwon¹, Dongchan Kim¹, Hyunseok Seo¹, Jaejin Cho¹, Byungjai Kim¹, and HyunWook Park^1
1KAIST, Daejeon, Korea, Republic of

A long imaging time has been regarded as a major drawback of MRI, and many techniques have been proposed to overcome this problem. Parallel imaging (PI) and compressed sensing (CS) techniques utilize different sensitivity of multi-channel RF coils and sparsity of signal in a certain domain to remove aliasing artifacts that are generated by subsampling, respectively. In this study, an artificial neural networks (ANN) are applied to MR reconstruction to reduce imaging time, and it is shown that the ANN model has a potential to be comparable to PI and CS.

Satoshi ITO¹, Shungo YASAKA¹, and Yoshifumi YAMADA^1
1Information and Controls Systems Sciences, Utsunomiya University, Utsunomiya, Japan

In this paper, we propose a new fast image reconstruction method in which a regularly undersampled signal is used instead of random sampling, as is used in compressed sensing. To diffuse the aliasing artifact caused by under-sampling, we adopt phase-scrambling Fourier transform imaging. The proposed method has an advantage over CS in that the quality of the image does not depend on the selection of the sampling trajectory. Simulation studies and experiments show that the proposed method has almost the same peak signal-to-noise ratio as that of a compressed sensing reconstruction.

Chia-Chu Chou¹, Taehoon Shin², JiaChen Zhuo², Gadisetti Chandramouli³, Murali Cherukuri³, and Rao Gullapalli^2
1Elecetrical Engineering, University of Maryland, College Park, Beltsville, MD, United States, ²1Diagnostic Radiology and Nuclear Medicine, University of Maryland School of Medicine, Baltimore, MD, United States,³National Cancer Institute, Bethesda, MD, United States

In Electron Paramagnetic Resonance Imaging (EPRI), each single k-space point is typically acquired per excitation and lengthens the acquisition time. In order to accelerate the imaging process, we raised a new image reconstruction method, Partial Fourier Compressed Sensing (PFCS), to address this problem. With PFCS, the images can be reconstructed from 25% of the k space and hence accelerate the imaging process to less than 1min. We also demonstrated PFCS reconstructed linewidth map were able to monitor the oxygen change in the tumor tissue.

Carlos Milovic^1,2, Jose Miguel Pinto^1,2, Julio Acosta-Cabronero³, Pablo Irarrazaval^1,2, and Cristian Tejos^1,2
1Department of Electrical Engineering, Pontificia Universidad Catolica de Chile, Santiago, Chile, ²Biomedical Imaging Center, Pontificia Universidad Catolica de Chile, Santiago, Chile, ³German Center for Neurodegenerative Deceases (DZNE), Magdeburg, Germany

An strategy for parameter selection in TGV regularized reconstruction problems is presented, with applications to deconvolutions and QSM. This allows fine-tuning of parameters in an efficient way and the use of predictors that are correlated to optimized results in terms of MSRE. This allows users to automatize or accelerate the parameter selection, critical in expensive problems such as QSM and reduce the error in the reconstruction.

Steven T. Whitaker¹, Meredith Taylor¹, Haonan Wang¹, and Neal K. Bangerter^1
1Electrical Engineering, Brigham Young University, Provo, UT, United States

Balanced steady-state free precession (bSSFP) provides high signal in short scan times.  A new method for combining four or more phase-cycled bSSFP acquisitions was recently proposed that uses an elliptical signal model (ESM) of the bSSFP signal.   In this study, we compare the SNR performance and effectiveness at reducing banding artifact of the ESM, complex sum, and sum-of-squares techniques across a range of T1 and T2 values, flip angles, and base SNR levels.   Although ESM produces near perfect band removal in high SNR situations, it breaks down for certain tissues and in low SNR situations.

Nao Kodama¹ and Katsumi Kose^1
1Institute of Applied Physics, University of Tsukuba, Tsukuba, Japan

Echo planar imaging sequences were developed for a 9.4 Tesla vertical standard bore (~54 mm) superconducting magnet using an unshielded gradient coil. Because EPI requires fast switching of intense magnetic field gradients, eddy currents were induced in the surrounding metallic materials, e.g., the room temperature bore, and this produced serious artifacts on the EPI images. We solved the problem using an unshielded gradient coil set of proper size (39 mm OD, 32 mm ID) and reference scans. The obtained EPI images of a phantom and a plant sample were almost artifact free and demonstrated the promise of our approach.

Keith R. Thulborn¹, Chao Ma², Ian C. Atkinson¹, Theodore C. Claiborne¹, Steven M. Wright³, and Reiner Umathum^4
1Center for Magnetic Resonance Research, University of Illinois, Chicago, IL, United States, ²Beckman Institute, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ³Department of Biomedical Engineering, Texas A&M University, College Station, TX, United States, ⁴Division of Medical Physics in Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany

SERIAL excitation produces uniform image intensity at low power at ultra high field but has not been applied to humans. FLASH sequences modified for sequential single coil excitation while retaining full array receive mode were combined with generalized total variation regularized SENSE reconstruction and 4- and 8- arrayed coils. Images with acceptable uniformity, contrast and resolution over the in vivo human brain are demonstrated at 9.4T using low power.

Wei Liu¹ and Kun Zhou^1
1Siemens Shenzhen Magnetic Resonance Ltd, Shenzhen, China, People's Republic of

In this study, an optimized slice acquisition ordering method was proposed to improve the signal attenuation and contrast alteration caused by the crosstalk and MT effect in multi-slice HASTE imaging using a short TR. It demonstrated that a shorter acquisition time is possible with an optimized slice acquisition order. It allows a shorter TR, whilst maintaining the SNR and contrast similar to the conventional one, which is particularly useful in the abdominal imaging.

Ali Caglar Özen¹, Jan Korvink², and Michael Bock^1
1Dept. of Radiology - Medical Physics, University Medical Center Freiburg, Freiburg, Germany, ²Institute of Microstructure Technology, Karlsruhe Institute of Technology, Karlsruhe, Germany

Concurrent Excitation and Acquisition (CEA) enables MRI with true zero echo times, and full signal acquisition efficiency. However, frequency sweep along readout gradients results in sequential excitation of spins at different locations, thus a unique TR is experienced by each spin at each radial acquisition spoke. In this work, implications of modulations in transverse magnetization as a function of T1, flip angle and TR were investigated for 2D and 3D radial acquisition schemes with equidistant point trajectory, segmented ordering and golden angle ordering. Resulting changes in point spread function (PSF) of a point source located at the edge of the field of view (FOV) were analyzed and discussed.

Neville D Gai¹, Ashkan A Malayeri¹, and David A Bluemke^1
1Radiology & Imaging Sciences, NIH, Bethesda, MD, United States

Lung T1/T2* may be useful in discriminating between normal and pathological tissue particularly in disorders such as fibrosis, edema or emphysema. Quantitative mapping of the lung parenchyma is challenging due to the low proton density, respiratory and cardiac motion and susceptibility effects. Here we describe a technique based on segmented respiratory triggered 3D ultrashort echo time dual-echo radial imaging interleaved with and without a WET saturation pulse to estimate T1 and T2* maps simultaneously in a single scan.  The results show that T1/T2* mapping of lung parenchyma can be reliably performed with relatively high resolution in a clinically feasible time.

Wingchi Edmund Kwok^1,2
1Department of Imaging Sciences, University of Rochester, Rochester, NY, United States, ²Rochester Center for Brain Imaging, Rochester, NY, United States

Ultrashort echo-time imaging has been explored for the study of short T2* tissues.  Most ultrashort TE sequences utilize 3D radial center-out k-space sampling.  While they are potentially useful for many important applications, they are susceptible to various image artifacts.  This abstract describes the appearances, causes and mitigations of some potential artifacts, which include those caused by long readout length, high gradient field, insufficient number of radial projections, off-centered imaging and signal wrap-around.  An understanding of these artifacts will help in protocol setting and the identification of related problems.  This article should benefit the users of ultrashort TE imaging.

Joshua M Kuperman¹, Nathan S White¹, Hauke Bartsch¹, Matthew Middione², Kun Lu³, Thomas Liu³, Terry Jernigan⁴, Ajit Shankaranarayanan², and Anders M Dale^1,5
1Radiology, University of California, San Diego, La Jolla, CA, United States, ²GE Healthcare, Menlo Park, CA, United States, ³Center for Functional Magnetic Resonance Imaging, University of California, San Diego, La Jolla, CA, United States, ⁴Center for Human Development, University of California, San Diego, La Jolla, CA, United States, ⁵Neurosciences, University of California, San Diego, La Jolla, CA, United States

The benefits of the advanced MRI protocols used in the Human Connectome Project have heretofore only been available on Siemens MRI scanners.  We have designed a Connectome-like protocol, called the Harmonized Human Connectome Protocol, which can utilize MRI scanners from additional vendors, specifically GE and Philips.  This protocol is particularly relevant for the Adolescent Brain and Cognitive Development Study, which aims to scan over 10,000 children ages 9-10 and follow them longitudinally for ten years.  This abstract details the protocol for the GE MR750 scanner and outlines calibration and correction procedures which can be used to further reduce unwanted site/scanner effects.

Jinbang Guo^1,2, Xuefeng Cao^1,3, Zackary I. Cleveland¹, and Jason C. Woods^1,2
1Center for Pulmonary Imaging Research, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States, ²Physics, Washington University in St. Louis, St. Louis, MO, United States, ³Department of Physics, University of Cincinnati, Cincinnati, OH, United States

Motion due to respiration is one of the major difficulties in lung imaging of mice, which have a 10-20-fold higher respiratory rate than humans.  In this study, we demonstrate that the FID signal amplitude (k = 0) as a function of projection number in center-out radial 3D UTE reflects respiratory motion. Retrospective "self"-gating using this FID signal amplitude was applied to extract data for end-expiration and end-inspiration respectively. Quantitative analysis of tidal volumes and lung parenchymal signal match external measurements and physiological expectations.

Muhammad Usman¹, Gastao Cruz¹, and Claudia Prieto^1
1Division of Imaging Sciences and Biomedical Engineering, King's College London, London, United Kingdom

In this work, we propose to use a novel free-running self-gated 3D CArtesian acquisition with Spiral PRofile ordering and TIny Golden angle step for Eddy Current Reduction, so called CASPR-TIGER. Data is acquired continuously under free breathing (no ECG gating, no pre-pulses interruption) using CASPR-TIGER trajectory and 4D volumes (3D+time) are reconstructed from all available data (100% respiratory scan efficiency) using a soft gating technique combined with temporal total variational (TV) constrained iterative SENSE reconstruction. Feasibility of proposed method is demonstrated in three subjects in a 3-3.6 minutes free breathing acquisition.

Aurelien J Trotier¹, Charles R CASTETS¹, William LEFRANCOIS¹, Emeline J RIBOT¹, Eric THIAUDIERE¹, Jean-Michel FRANCONI¹, and Sylvain MIRAUX^1
1RMSB-UMR5536, CNRS - Université de Bordeaux, Bordeaux, France

4D flow MRI on mouse models remains very difficult due to the very small size of vessels and the extremely high cardiac rhythm. To overcome this problem a 3D time-resolved Phase Contrast UTE sequence was combined with an injection of Ultra Small Particles of Iron Oxide to obtain a positive and high signal in blood. The method was exploited to quantify blood flow velocity of the cardiovascular system in mice with a high spatial  (200 µm)3 and temporal resolution (16ms). The total acquisition can be reduced to 25min by limiting the number of acquired projections per cine image.

Daniel Gallichan^1
1CIBM, EPFL, Lausanne, Switzerland

We extended previous work using 3D-FatNavs to enable motion-correction of ultra-high resolution structural acquisitions, including T1-, T2- and T2*-weighted images. Images are of exceptional quality and detail for in-vivo acquisitions.

Hernan Jara¹, Arnaud Guidon², Jorge A Soto¹, and Osamu Sakai^1
1Radiology, Boston University, Boston, MA, United States, ²Global MR Applications and Workflow, GE Healthcare, Boston, MA, United States

Purpose: To describe the quadra fast spin-echo (quadra-FSE) pulse sequence, which is the concatenation of two dual-echo FSE acquisitions differing only in TR and to describe the matching qMRI algorithms for mapping T1, T2, and PD. Methods: quadra-FSE was tested at 3T with a multi-compartment agarose phantom and relaxometry was compared to gold standard relaxometry scans using qMRI algorithms developed in house. Results: PD, T1, and T2 maps generated with the quadra-FSE scans are accurate and of excellent image quality. Conclusion: Concatenation of two DE-FSE scans with different TRs can be used for combined and accurate PD, T1, and T2 mapping.

Zechen Zhou¹, Shuo Chen¹, Aiqi Sun¹, Yunduo Li¹, Rui Li¹, and Chun Yuan^1,2
1Center for Biomedical Imaging Research, Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing, China, People's Republic of, ²Vascular Imaging Laboratory, Department of Radiology, University of Washington, Seattle, WA, United States

A parametric variable radius sampling scheme termed Cartesian Under-Sampling with Target Ordering Method (CUSTOM) was introduced for undersampling pattern design to better match the total number of sampling points with the given acceleration factor in 3D Cartesian imaging application. With the same joint parallel imaging and compressed sensing image reconstruction method, parameter optimized CUSTOM has demonstrated its enhanced performance particularly for detail image information restoration in comparison to several undersampling pattern design schemes, as well as its generalization ability in different applications. The prospective experiment validated the feasibility of CUSTOM in clinical settings.

Jana Maria Hutter¹, Anthony N Price¹, Lucilio Cordero Grande¹, Emer Judith Hughes¹, Kelly Pegoretti¹, Andreia Oliveira Gaspar¹, Laura McCabe¹, Mary Rutherford¹, and Joseph V Hajnal^1
1Centre for the developing brain, King's College London, London, United Kingdom

Quiet sequences are of particular importance for fetal EPI based imaging, where the necessary protection of the unborn infant can often compromise the efficiency and achievable resolution of the EPI acquisition. This is of particular relevance for connectome type studies, where long functional and diffusion weighted sequences need to be acquired in an efficient and safe way.This abstract presents a quiet SE and GE EPI framework with sinusoidal read-out constant phase and merged crusher strategy, completely flexible and adaptable to the scanner impulse response function leading to a decrease of up to 9dB(A).

David Andrew Porter¹ and Marco Vicari^1
1Fraunhofer MEVIS, Bremen, Germany

A novel method of echo-planar spectroscopic imaging is introduced, in which readout segmentation is used to reduce the echo spacing and provide a substantial increase in spectral bandwidth. Results are presented, showing how the technique avoids the aliasing problems that affect conventional applications of high-resolution, spectroscopic imaging at 3T and serves as a robust method for providing spectrally-selective fat and water images. The method is also a promising option for high-bandwidth, spectroscopic imaging studies of metabolites at high field strengths. 

Kun Zhou¹ and Wei Liu^1
1Siemens Shenzhen Magnetic Resonance Ltd., Shenzhen, China, People's Republic of

A turbo spin echo based sequence for robust and fast diffusion imaging is proposed. It overcomes the non-CPMG problem by split-echo acquisition of turbo spin echo signals. EPI-like readout is used to sample the separated echoes and generate multiple blades for a single k-space. Each blade was corrected for both the inherent phase of separated echoes and off-resonance phase, to avoid the destructive inference. With this technique, the non-CPMG problem can be effectively mitigated at low flip angle refocusing pulses to reduce SAR. Moreover, the off-resonance artifacts can also be reduced especially when high acceleration factor is applied.

Jiasheng Su¹ and Shaoying Huang^1
1EPD, Singapore University of Technology and Design, Singapore, Singapore

For a low-field MRI system, the inverse calculation of the encoding matrix is time consuming and moreover, there is a blurry area at the center of the reconstructed image. To solve this problem, three strategies are proposed. Firstly, QR decomposition is applied to inverse the matrix to eliminate the blurry area. Secondly, the encoding matrix is separated so that  the results of the matrix inverse can be reused. Last, the size of encoding matrix is reduced by optimizing sample points. One example is given, the calculation time is reduced, and the imaging quality is improved. The proposed approach increases the imaging capability of a low-field MRI system. 

Ludovic de Rochefort¹, Geneviève Guillot¹, Rose-Marie Dubuisson¹, and Romain Valabrègue^2
1Imagerie par résonance magnétique médicale et multi-modalités, IR4M, UMR 8081, CNRS-Université Paris-Sud, Université Paris-Saclay, Orsay, France, ²CENIR, ICM, Inserm U 1127, CNRS UMR 7225, Sorbonne Universités, UPMC Univ Paris 06 UMR S 1127F, Paris, France

Fast steady-state sequences combine RF and gradient spoiling to modulate contrasts in MRI. The steady-state depends on many physical and acquisition parameters. Here, in vivo feasibility on brain is shown to map proton density, background phase, flip angle, relaxation rates and apparent diffusion coefficient from such sequences. Multiple volumes were acquired with various optimized prescribed flip angle, spoiling gradients and phase increments, and the complex signal was fitted to the Bloch-Torrey signal model with free diffusion using efficient calculation algorithms. The acquisition of full 3D co-localized multi-parametric maps of relevant MR physical parameters in a realistic scan time is demonstrated.

Albrecht Ingo Schmid^1,2, Martin Meyerspeer^1,2, Simon Daniel Robinson^2,3, Martin Krssak^2,3,4, Michael Wolzt⁵, Ewald Moser^1,2, and Ladislav Valkovic^2,3
1Center for Medical Physics and Biomedical Engineering, Medical University of Vienna, Vienna, Austria, ²MR Centre of Excellence, Medical University of Vienna, Vienna, Austria, ³Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria, ⁴Department of Internal Medicine 3, Medical University of Vienna, Vienna, Austria, ⁵Department of Clinical Pharmacology, Medical University of Vienna, Vienna, Austria

MR spectroscopy provides valuable information about tissue metabolism but suffers from slow acquisition or poor spatial resolution and coverage. PCr and pH kinetics are the two important quantities derived from³¹P MR data. MRI has been used to measure PCr in the past, but not pH. Simultaneous fast 3-D gradient-echo images of PCr and P_i were acquired in healthy volunteers at 7T during exercise recovery. pH was calculated from phase images. Results of PCr and pH kinetics are comparable to MRS data. In conclusion, ³¹P MRI is an alternative to ³¹P MRS for fast coverage of multiple ROIs and low SAR.

Paul Kinchesh¹, Stuart Gilchrist¹, Ana L Gomes¹, Veerle Kersemans¹, John Beech¹, Danny Allen¹, and Sean Smart^1
1Department of Oncology, University of Oxford, Oxford, United Kingdom

We demonstrate that cardio-respiratory synchronisation can be achieved in conjunction with short TR scans and k-space segmentation to reduce imaging times to below that achievable using standard techniques such as retrospective gating.  Our method is generally applicable to other short TR scan modes requiring cardio-respiratory synchronisation. Images of the mouse heart, lung and liver are presented for the b-SSPF scan mode.

Hyungseok Jang^1,2 and Alan B McMillan^1
1Department of Radiology, University of Wisconsin, Madison, WI, United States, ²Department of Electrical and Computer Engineering, University of Wisconsin, Madison, WI, United States

Accurate MR-based attenuation correction (MRAC) is necessary to enable quantitative PET imaging in PET/MR. Unfortunately, identification of bone via MR methods is technically challenging due to its short T2*. Thus, ultrashort time echo (UTE) techniques have been proposed. In this study, we explored rapid multi-echo frequency encoded UTE and ramped hybrid encoding (RHE) for UTE-based imaging schemes for MRAC with clinically feasible scan times (<35sec). By using an IDEAL-based signal model for long T2* suppression, multi-echo hybrid encoding UTE imaging performed better than frequency encoded UTE. 

André Fischer^1,2, Michael N. Hoff³, Piero Ghedin^1,2,4, and Anja C.S. Brau^2
1GE Global Research, Garching bei München, Germany, ²Cardiac Center of Excellence, GE Healthcare, Garching bei München, Germany, ³Department of Radiology, University of Washington, Seattle, WA, United States, ⁴GE Healthcare, Waukesha, WI, United States

Banding artifacts in bSSFP sequences pose a challenge in cardiac cine imaging, especially at 3.0T. Recently, a “Geometric Solution” (GS) which is capable of completely removing banding artifacts has been introduced and demonstrated in applications outside the heart. This work investigates the feasibility of extending GS to cardiac cine imaging at 3.0T and explores its potential to enable longer TRs than have conventionally been feasible with bSSFP, permitting sub-millimeter resolution cine imaging free of banding artifacts.

Tom Hilbert^1,2,3, Tobias Kober^1,2,3, Jean-Philippe Thiran^2,3, Reto Meuli², and Gunnar Krueger^2,3,4
1Advanced Clinical Imaging Technology (HC CMEA SUI DI BM PI), Siemens Healthcare AG, Lausanne, Switzerland, ²Department of Radiology, University Hospital (CHUV), Lausanne, Switzerland, ³LTS5, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, ⁴Siemens Medical Solutions USA, Inc., Boston, MA, United States

Phase-encode ghosting artifacts frequently occur in magnetic resonance imaging, especially in spin-echo sequence derivatives such as fluid-attenuated inversion recovery. The appearance of these artifacts may cause misinterpretation as tissue pathology, e.g. a lesion. We propose an algorithm to automatically detect these artifacts by analyzing the consistency of the acquired k-space with respect to the assumption of GRAPPA that a k-space sample is a linear sum of its neighboring samples. The performance of the technique is shown in three volunteers. It may help to avoid potential misinterpretation in the future, both for radiological readers and automated post-processing algorithms.

Hongpyo Lee¹, Yoonho Nam², Min-Oh Kim¹, Dongyeob Han¹, and Dong-Hyun Kim^1
1School of Electrical and Electronic Engineering, Yonsei University, Seoul, Korea, Republic of, ²Department of Radiology, Seoul St. Mary Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea, Republic of

Recently, myelin water fraction was investigated using multi-echo GRE data. Generally, to ensure phase consistency among the echoes, multi-echo acquisitions use unipolar gradients. However, these unipolar gradient multi-echo sequences reduce acquisition efficiency and increase echo spacing. Bipolar gradients would be preferential however, k-space misregistration induced by readout gradient delays and eddy-currents make phase errors, so severe artifacts occur in myelin water imaging. In this abstract, we present a MWI using bipolar gradient multi-echo GRE sequence with k-space shift correction. Compared to unipolar MWF, k-space shift corrected bipolar MWF yields a reduction in ΔTE, which leads to improved SNR and more accurate quantification.

Scott D. Swanson¹, Dariya I. Malyarenko¹, and Thomas L. Chenevert^1
1Department of Radiology, University of Michigan, Ann Arbor, MI, United States

Quantitative temperature mapping

Jacob Macdonald¹, Oliver Wieben^1,2, Scott K Nagle², and Kevin M Johnson^1
1Medical Physics, University of Wisconsin - Madison, Madison, WI, United States, ²Radiology, University of Wisconsin - Madison, Madison, WI, United States

Streaking artifacts in radial acquisitions from undersampling or data inconsistencies can reduce SNR and make it difficult to discern features in low signal areas. Anatomy that is outside of the imaging volume of interest but within the excitation volume can contribute to these artifacts. We used out-of-volume spatial saturation pulses to suppress these streaking artifacts with minimal scan time penalties. In-vivo acquisitions with spatial saturation showed equal or superior quality in all cases. They should be implemented whenever the additional SAR can be tolerated.

Patrick Alexander Liebig^1,2, Robin Martin Heidemann², and David Andrew Porter^3
1Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany, ²Siemens Healthcare GmbH, Erlangen, Germany, ³Fraunhofer MEVIS, Bremen, Germany

A new approach to Echo-Planar Imaging (EPI) is introduced under the name Zigzag-Aligned-Projections (ZAP) that replaces the blipped phase-encoding (PE) gradient with the modulus of the readout (RO). This comes with two significant advantages: the reduction of acoustic noise due to the modified PE gradient and the higher efficiency due to continuous data sampling. ZAP EPI is the only EPI derivate that combines Cartesian GRAPPA using a fixed Kernel size with continuous data sampling. The reduction in acoustic noise was verified experimentally and volunteer images were acquired and processed with two reconstruction techniques.

Zhiqiang Li¹, Dinghui Wang¹, John P Karis², and James G Pipe^1
1Imaging Research, Barrow Neurological Institute, Phoenix, AZ, United States, ²Neuroradiology, Barrow Neurological Institute, Phoenix, AZ, United States

Turbo spin-echo (TSE) provides rapid T2-weighted imaging with slightly altered contrast compared to Cartesian spin-echo (SE). A spiral SE technique has been proposed for fast T2-weighted imaging without degrading the T2 contrast. In this study a dual echo spiral SE with hybrid spiral readouts is developed to simultaneously provide both proton density and T2 contrast without increasing the scan time. Volunteer results from the spiral dual SE technique demonstrate similar contrast to conventional SE, with a scan speed faster than Cartesian mDixon TSE.

Nicholas R. Zwart¹, Dinghui Wang¹, and James G. Pipe^1
1Imaging Research, Barrow Neurological Institute, Phoenix, AZ, United States

Time of Flight MRA in the head and neck can benefit from Dixon fat signal removal.  The spiral trajectory is used to speed up the acquisition allowing 3-echoes to be collected, for Dixon reconstruction, in less time than a conventional single echo ("out-of-phase") cartesian MRA.

Lennart Geurts¹, Sander Brinkhof¹, Peter R. Luijten¹, and Jaco J.M. Zwanenburg^1
1Radiology, UMCU, Utrecht, Netherlands

Because cerebral perforating arteries have sub-millimeter diameters and slow blood flow velocities, their blood flow velocity and pulsatility measurements are challenging and limited by noise and partial volume effects. Our previously reported acquisition method used two signal averages (NSA) to increase the signal-to-noise ratio (SNR). We show that decreasing NSA, and thereby reducing scan time by half, has little effect on vessel detection. The NSA=1 coefficients of repeatability (CoR) found in this study are similar to previously published NSA=2 CoR`s. Subject motion and small vessel size likely play together to cause a sub-optimal benefit from increased imaging time.

Xiaojuan TAO¹, Yun PENG¹, Yanqiu LV¹, and Kaining SHI^2
1Imaging Center of Beijing Children's Hospital Affiliated To Capital Medical University, Beijing, China, People's Republic of, ²Imaging Systems Clinical Science Philips Healthcare, Beijing, China, People's Republic of

Spleen and liver are common organs involved in Gaucher disease (GD), while few reports have been published on the measurement of  volume and lipid content of liver and spleen using MR. This study recruited 42 patients with GD by 1.5T MR. Our results showed that  enlargement of spleen was more severe than liver. Spleen has higher lipid content than liver among these patients, while both of them exhibited higher fat fraction than normal value. Our study suggests MRI can be employed to monitor the disease progression and effect of the treatment in children and adolescent patients with GD.

Rahel Heule^1,2, Philipp Madörin^1,2, and Oliver Bieri^1,2
1Division of Radiological Physics, Department of Radiology, University of Basel Hospital, Basel, Switzerland, ²Department of Biomedical Engineering, University of Basel, Basel, Switzerland

While the steady-state free precession (SSFP) dynamics of spin-1/2 nuclei such as ¹H obey the Bloch equations, a similar mathematical framework for describing and understanding the characteristics of ²³Na SSFP signal behavior is not yet available. In this work, sodium MRI probes were investigated and, in particular, a novel class of phantoms was presented that proved the ability to generate high sodium signal at 3 Tesla without impairment due to the skin effect. By means of balanced SSFP frequency profile measurements, the potential of the novel phantoms to explore ²³Na SSFP was demonstrated.

Tim Hilgenfeld¹, Alexander Heil¹, Sebastian Schwindling², David Grodzki³, Mathias Nittka³, Daniel Gareis⁴, Peter Rammelsberg², Martin Bendszus¹, Sabine Heiland¹, and Marcel Prager^1
1Division of Neuroradiology, University Heidelberg, Heidelberg, Germany, ²Division of Prosthodontics, University Heidelberg, Heidelberg, Germany, ³Siemens Healthcare GmbH, Erlangen, Germany, ⁴NORAS MRI products GmbH, Höchberg, Germany

Dental MRI is a new technique which is often impaired by artifacts due to metallic dental implants. Several MRI sequences were developed to reduce susceptibility artifacts (e.g. for orthopaedic implants). Here, we for the first time systematically evaluated MR sequences for artifact reduction in dental implants. Smallest artifact volume was measured for 2D-TSE sequences. Since imaging of dental structures benefit from high resolution and possibility of 3D reconstructions 3D sequences are advantageous. Significant artifact reduction was noted for SPC-WARP measuring only 2.1 times artifact volume of TSE sequence instead of 4.8 times when using standard SPC sequence.

Aaron T Hess¹, Christopher T Rodgers¹, and Matthew D Robson^1
1OCMR, University of Oxford, Oxford, United Kingdom

The reflected power of transmit RF coils is influenced by the position of the diaphragm. In this work the diaphragm position is measured in real-time for every RF pulse with a hybrid approach. The set of reflection coefficients are transformed into a diaphragm position using a series of MR diaphragm navigators at the start of the pulse sequence in a learning cycle. We demonstrate high quality respiratory gated data based on gating via this mechanism using standard SAR monitoring hardware with a real-time lag of 23ms and temporal resolution of 4.5ms.

Guobin Li¹, Zhaopeng Li¹, Chaohong Wang¹, Yang Xin¹, Shuheng Zhang¹, Weijun Zhang¹, Xiaodong Zhou¹, and Weiguo Zhang^1
1Shanghai United Imaging Healthcare Co., Ltd, Shanghai, China, People's Republic of

C-spine imaging is demanding due to artifacts from CSF flow, patient’s swallowing etc., especially in FSE sequence with inversion recovery. By increasing the excitation thickness in FSE sequence, two concomitant saturation bands are realized at both sides of each image slice, which suppress the moving CSF. Furthermore, a snapshot k-space ordering is proposed to further improve the stability of c-spine imaging against irregular flow of CSF and patient’s bulk motion.

Mootaz Eldib¹, Li Feng², Daniel K Sodickson², Zahi A Fayad¹, and Hadrien A Dyvorne^1
1Translational and Molecular Imaging Institute, Icahn school of Medicine at Mount Sinai, New York, NY, United States, ²Center for Advanced Imaging Innovation and Research (CAI2R), Bernard and Irene Schwartz Center for Biomedical Imaging, Department of Radiology, New York University School of Medicine, New York, NY, United States

We propose a novel acquisition technique for motion-resolved abdominal imaging. Using a golden angle spiral projection trajectory, we were able to acquire reliable physiologic tracking data while acquiring 3D isotropic resolution images of the entire upper abdomen, resulting in an efficient self-gated sequence. We show that respiratory motion can be fully characterized in vivo in a minute-long 1.8 mm isotropic acquisition, which is suitable for applications such as PET/MR motion correction.

Jieying Luo¹, Nii Okai Addy¹, R. Reeve Ingle¹, Corey A. Baron¹, Joseph Y. Cheng¹, Bob S. Hu^1,2, and Dwight G. Nishimura^1
1Electrical Engineering, Stanford University, Stanford, CA, United States, ²Palo Alto Medical Foundation, Palo Alto, CA, United States

3D image-based navigators (iNAVs) offer the potential to achieve more complete motion correction for coronary magnetic resonance angiography (CMRA). In this work, we develop a method for 3D-iNAV processing to achieve nonrigid motion correction. Both global and localized motion trajectories are extracted from the 3D iNAVs and used to generate candidate motion-corrected images for an autofocus method. Two sets of localized motion trajectories are obtained from deformation fields between 3D iNAVs and reconstructed binned images respectively. Results with this method on whole-heart 3D cones CMRA scans demonstrated improved vessel sharpness as compared to 3D translational motion correction.

Onur Afacan¹, Burak Erem¹, Diona P. Roby¹, Noam Roth², Amir Roth², Sanjay P. Prabhu¹, and Simon K. Warfield^1
1Radiology, Boston Childrens Hospital and Harvard Medical School, Boston, MA, United States, ²Robin Medical Inc., Baltimore, MD, United States

In this work we report results from a large pediatric study that shows the effect of motion. Motion patterns were measured on 82 children, mean age 13.4 years, in a T1 weighted brain MRI. An expert radiologist graded the images using a 4-point scale ranging from clinically non-diagnostic to no motion artifacts. We used these grades to correlate motion parameters such as maximum motion, mean displacement from a reference point and motion free time. The results will help the motion correction community in better understanding motion patterns in pediatric populations and how it effects image quality.

Daniel Christopher Hoinkiss¹, Matthias Guenther¹, and David Andrew Porter^1
1MR Physics, Fraunhofer MEVIS, Bremen, Germany

Diffusion Tensor Imaging is frequently affected by long-term subject motion. Intermediate pseudo-trace-weighted images enable a real-time image registration with low sensitivity to contrast variation between diffusion-weighted images. These registration results are used to correct the imaging parameters of the ongoing scan. The algorithm was evaluated on three individual subjects using a dedicated diffusion-weighted imaging sequence. The prospective motion correction was able to reduce the typical long-term motion to a band of approximately ±0.2mm for translational and ±0.2° for rotational motion, which is far below voxel size, without increasing the total scan time or changing the set of diffusion vectors.

Martin Schwartz^1,2, Thomas Küstner^1,2, Christian Würslin¹, Petros Martirosian¹, Nina F. Schwenzer³, Fritz Schick¹, Bin Yang², and Holger Schmidt^3
1Section on Experimental Radiology, Department of Radiology, University of Tuebingen, Tuebingen, Germany, ²Institute of Signal Processing and System Theory, University of Stuttgart, Stuttgart, Germany, ³Diagnostic and Interventional Radiology, Department of Radiology, University of Tuebingen, Tuebingen, Germany

Respiratory motion-free images are important in MRI of the human thorax and abdomen. A significant factor is the reconstruction of these images for the application in clinical practice. The objective of the presented work is an integration of an approved motion correction algorithm into the clinical environment to overcome limitations of offline reconstructed images by the utilization of external workstations. Therefore, a reconstruction pipeline based on the open-source framework Gadgetron with new modules for the integration of a motion correction algorithm is demonstrated.

Alexander Loktyushin^1,2, Philipp Ehses¹, Bernhard Schölkopf², and Klaus Scheffler^1,3
1High-field Magnetic Resonance, Max Planck Institute for Biological Cybernetics, Tübingen, Germany, ²Empirical Inference, Max Planck Institute for Intelligent Systems, Tübingen, Germany, ³Biomedical Magnetic Resonance, University of Tübingen, Tübingen, Germany

Long-TE gradient-echo images are prone to ghosting artifacts. Such degradation is primarily due to magnetic field variations caused by breathing or motion. The effect of these fluctuations amounts to different phase offsets in each acquired k-space line. A common remedy is to measure the problematic phase offsets using an extra non-phase-encoded scan before or after each imaging readout. In this work, we attempt to estimate the phase offsets directly from the raw image data by optimization-based search of phases that minimize an image distortion measure. This eliminates the need for any sequence modifications and additional scan time.

Isabella Radl¹, Stephen Keeling², and Rudolf Stollberger^1,3
1Institute of Medical Engineering, Graz University of Technology, Graz, Austria, ²Institute for Mathematics and Scientific Computing, Karl Franzens University of Graz, Graz, Austria, ³BioTechMed Graz, Graz, Austria

Different applications in DCE-MRI suffer from inter-frame misalignment due to physiological motion, which has to be compensated for further analysis of functional parameters. Conventional motion correction methods are usually unable to register images with simultaneous changes of contrast and morphology. Virtual-template based registration overcomes this problem by iteratively generating a motion-less image series with the contrast behaviour of the original DCE data as registration targets. We investigated different methods to generate these virtual-templates and identified Independent Component Analysis as best approach among the investigated techniques. Results were validated on a synthetic kidney phantom and in-vivo myocardial perfusion MRI.

Patricia M Johnson^1,2, Junmin Liu¹, Trevor Wade¹, and Maria Drangova^1,2
1Robarts Research Institute, London, ON, Canada, ²Department of Medical Biophysics, Schulich School of Medicine & Dentistry, London, ON, Canada

Spherical navigators are k-space navigators that can measure 6-degree of freedom rigid-body motion. Recent developments have reduced processing and baseline acquisition time, making the technique a promising tool for motion correction.  This work represents the first time SNAVs have been incorporated into an image sequence and demonstrated for motion correction. SNAVs were incorporated into a gradient echo sequence; this navigated sequence was used to scan 3 volunteers performing directed head motion. The motion-degraded brain images were then retrospectively corrected using the SNAV derived motion parameters. In all cases excellent correction of motion artifacts was observed. 

Mario O. Malavé¹, Nii Okai Addy¹, R. Reeve Ingle¹, Joseph Y. Cheng¹, Corey A. Baron¹, and Dwight G. Nishimura^1
1Electrical Engineering, Stanford University, Stanford, CA, United States

Motion and eddy current artifacts were investigated with simulations, metric measures, and in vivo scans for three different cone acquisition schemes: sequential, multidimensional golden means (MDGM), and phyllotaxis readout orderings. We demonstrate the idea of using the 3D cones phyllotaxis acquisition method for improved motion behavior and low eddy current susceptibility. Also, the sequential ordering method is shown to be more susceptible to motion artifacts while the MDGM introduces eddy current artifacts. When using the phyllotaxis design, the reconstruction demonstrates that a more spread out k-space traversal per heartbeat is more robust to motion and can be obtained without introducing eddy currents.

Xinwei Shi^1,2, Joseph Y Cheng^1,2, Michael Lustig³, John M Pauly², and Shreyas S Vasanawala^1
1Radiology, Stanford University, Stanford, CA, United States, ²Electrical Engineering, Stanford University, Stanford, CA, United States, ³Electrical Engineering and Computer Science, UC Berkeley, Berkeley, CA, United States

In cardiac MRI, self-gating using the DC signal provides a promising alternative to EKG gating. However, the DC signal is often affected by other moving structures in the FOV, such as the liver, which degrades the accuracy of the extracted cardiac triggers. In this work, we demonstrate the use of virtual coils to focus the DC signal on cardiac motion and to provide a robust and generalized self-gating approach for 3D cardiac MRI. In free-breathing 4D-Flow scans of pediatric subjects, the proposed method improved the accuracy of self-gating triggers, and the self-gated images showed comparable quality with EKG gated reference. 

Guobin Li¹, Zhaopeng Li¹, Chaohong Wang¹, Yang Xin¹, Weijun Zhang¹, Xiaodong Zhou¹, and Weiguo Zhang^1
1Shanghai United Imaging Healthcare Co., Ltd, Shanghai, China, People's Republic of

To reduce pulsatile artifacts of blood flow in FSE imaging, a combined solution is proposed, in which two concomitant saturation bands are achieved at both side of each slice without any extra RF pulses and gradients. Furthermore, through a proper setting of slice acquisition order, flowing blood can be continuously and seamlessly saturated in multi-slice acquisition.

Hirofumi Hata¹, Yusuke Inoue², Ai Nakajima¹, Shotaro Komi¹, Yutaka Abe¹, Keiji Matsunaga², and Hiroki Miyatake^1
1Department of Radiology, Kitasato University Hospital, Sagamihara, Japan, ²Department of Diagnositic Radiology, Kitasato University School of Medicine, Sagamihara, Japan

We compared navigator-gating techniques for free-breathing 2D SPGR images of the liver using pencil-beam excitation and self-navigation techniques in 3 T MRI. In pencil-beam navigator, single-check (PB-SC) and double-check (PB-DC) modes were examined. In self-navigator scans, self-navigator signals were acquired in two fashions; before (SN-Pre) or after (SN-Post) the imaging read-out. Visual analysis shows that respiratory waveforms fluctuated in SN-Post. Quantitative and qualitative image evaluations show that PB-DC and SN-Post had better image qualities than the others. Considering scan time was about doubled in SN-Post, PB-DC should be the best for respiratory navigation in 2D SPGR imaging at this stage.

Yuji Iwadate¹, Atsushi Nozaki¹, Yoshinobu Nunokawa², Shigeo Okuda³, Masahiro Jinzaki³, and Hiroyuki Kabasawa^1
1Global MR Applications and Workflow, GE Healthcare Japan, Hino, Tokyo, Japan, ²Department of Radiation Technology, Keio University Hospital, Tokyo, Japan, ³Department of Diagnostic Radiology, Keio University School of Medicine, Tokyo, Japan

The conventional pencil-beam navigator suffers from large acoustic noise due to oscillating gradient pulses during RF excitation. We developed a silent navigator technique with whole volume excitation (vNav). The vNav technique reduced acoustic noise to almost the same level as background. In volunteer scan, a waveform of vNav was well correlated with the bellows signal, and motion reduction was demonstrated in 3D-SPGR imaging. The vNav integration into the silent imaging sequence should be examined in the next step.

Feiyu Chen¹, Tao Zhang^1,2, Joseph Y. Cheng^1,2, John M. Pauly¹, and Shreyas S. Vasanawala^2
1Electrical Engineering, Stanford University, Stanford, CA, United States, ²Radiology, Stanford University, Stanford, CA, United States

In this work, we propose a motion-robust auto-calibrating Wave-CS technique. This technique uses the wave-encoded center k-space and the known point-spread-function (PSF) of wave-encoding to reconstruct a Cartesian central k-space for calibration. The coil sensitivity maps are subsequently estimated with ESPIRiT for CS-SENSE reconstruction of under-sampled k-spaces. Results show this approach can reduce the motion artifacts and the aliasing artifacts due to sensitivity variations between the calibration and accelerated wave-encoded acquisitions.

Julian Maclaren¹, Andre Kyme^2,3, Murat Aksoy¹, and Roland Bammer^1
1Department of Radiology, Stanford University, Stanford, CA, United States, ²Department of Biomedical Engineering, University of California Davis, Davis, CA, United States, ³Brain and Mind Centre, University of Sydney, Sydney, Australia

Prospective motion correction based on optical tracking shows promise for improving image quality in MR brain imaging. To simplify this technique and expedite clinical deployment, it is desirable to avoid attaching markers to the patient’s head. Here we demonstrate proof-of-principle markerless tracking using an MR-compatible stereo camera and head coil configuration. We tested the method outside the MR environment using a 6-axis robot, capable of very accurate and repeatable (~20 µm) motion, to control a head phantom. Close agreement between our pose estimates and the applied motion suggests that accurate markerless tracking of the head is feasible in MRI. 

Seiichiro Noda¹, Nobuyuki Toyonari¹, Yukari Horino¹, Masami Yoneyama², and Kazuhiro Katahira^1
1Kumamoto Chuo Hospital, Kumamoto, Japan, ²Philips Electronics Japan, Tokyo, Japan

Gadoxetic acid enhanced mDIXON liver MRI has excellent utility for diagnosing hepatocellular carcinoma; however, it often suffers from ghosting flow artifacts from aorta due to increased signal by contrast enhancement. To solve this problem, we attempt to use motion-sensitized driven-equilibrium (MSDE) for reducing through-plane flow artifacts particularly in dynamic contrast-enhanced studies. We showed the effect of MSDE in reducing through-plane flow artifacts particularly in dynamic contrast-enhanced studies. Two types MSDE schemes (MSDE and iMSDE) could significantly decrease flow signals and could therefore reduce flow artifacts sufficiently. In current sequence, iMSDE would be better for clinical studies because of its less sensitivity to field inhomogeneities. 

Claus Benjaminsen¹, Rasmus Ramsbøl Jensen¹, Paul Wighton², M. Dylan Tisdall², Helle Hjorth Johannesen³, Ian Law³, Andre J. W. van der Kouwe², and Oline Vinter Olesen^1
1DTU Compute, Technical University of Denmark, Lyngby, Denmark, ²Athinoula. A. Martinos Center for Biomedical Imaging, Dept. of Radiology, Massachusetts General Hospital, Boston, MA, United States,³Department of Clinical Physiology, Nuclear Medicine & PET, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark

Prospective motion correction for MRI neuroimaging has been demonstrated using MR navigators and external tracking systems using markers. The drawbacks of these two motion estimation methods include prolonged scan time plus lack of compatibility with all image acquisitions, and difficulties validating marker attachment resulting in uncertain estimation of the brain motion respectively. We have developed a markerless tracking system, and in this work we demonstrate the use of our system for prospective motion correction, and show that despite being computationally demanding, markerless tracking can be implemented for real time motion correction.

Nicolai Spicher¹, Stefan Maderwald², Mark E. Ladd^2,3, and Markus Kukuk^1
1University of Applied Sciences and Arts Dortmund, Dortmund, Germany, ²Erwin L. Hahn Institute for Magnetic Resonance Imaging, University Duisburg-Essen, Essen, Germany, ³Division of Medical Physics in Radiology, German Cancer Research Center, Heidelberg, Germany

Videos of the human skin exhibit a subtle photoplethysmography signal, which resembles the one measured by pulse oximetry. It was investigated whether the whole photoplethysmography waveform (systolic/diastolic peak, dicrotic notch) can be extracted from two MR-compatible video cameras: A low-speed camera (30 frames-per-second) and a high-speed prototype (250 frames-per-second). We propose a potentially real-time feasible algorithm for signal filtering, which was applied to frames of both cameras. Using pulse oximetry as ground truth, revealed all features of the photoplethysmography waveform. Additionally, performing systolic peak detection showed that the high-speed camera allows for more accurate results in MRI pulse triggering.

Tim Nielsen¹ and Peter Börnert^1
1Philips Research, Hamburg, Germany

We present a reconstruction method to correct retrospectively for motion artifacts. The method identifies which part of the data set is affected by motion based on redundancy which is typically present in a multi-coil data set. No prior knowledge about coil sensitivity maps is needed. Instead, this information is directly estimated from the data along with the motion corrected image.

Thomas Küstner^1,2, Verena Neumann², Martin Schwartz^1,2, Christian Würslin^1,3, Petros Martirosian¹, Sergios Gatidis¹, Nina F. Schwenzer¹, Fritz Schick¹, Bin Yang², and Holger Schmidt^1
1Department of Radiology, University Hospital Tübingen, Tübingen, Germany, ²Institute of Signal Processing and System Theory, University of Stuttgart, Stuttgart, Germany, ³University of Stanford, Palo Alto, CA, United States

Motion estimation is an important task in MRI. For retrospective motion correction, there is often an image-based registration involved. Hence, the extraction of a reliable and accurate motion model for the underlying application is mainly dependent on the chosen image registration procedure. There are several different image registration methods available, but visualization and evaluation of the derived displacement fields and transformed images often remains an open topic. In the spirit of a reproducible research and for streamlining and simplifying the process, we provide GUIs and evaluation methods to perform and analyze image registration techniques which will be made publicly available.

Shufang Liu^1,2,3, Lin Zhang^3,4, Pauline Ferry^3,4, Andrei Codreanu⁵, Anne Menini², and Freddy Odille^3,4
1Institut für Informatik, Technology University of Munich, Munich, Germany, ²GE Global Research, Munich, Germany, ³Imagerie Adaptative Diagnostique et Interventionnelle, Université de Lorraine, Nancy, France,⁴U947, INSERM, Nancy, France, ⁵Centre Hospitalier de Luxembourg, Luxembourg, Luxembourg

This work present a robust motion correction framework for T1 mapping and ECV mapping. Motion correction within one series and between different series are discussed. Validation is performed on 2 patient data and 6 volunteer data.

Lirong Yan¹, Mayank Jog¹, Kay Jann¹, Xingfeng Shao¹, and Danny JJ Wang^1
1Neurology, University of California Los Angeles, Los Angeles, CA, United States

The alternations of brain pulsatility are related to various pathological changes such as traumatic brain injury (TBI) and brain tumor. In the present study, we introduced a new MRI approach to assess the brain’s biomechanical features using ECG-gated cine 3D integrated-SSFP, which offers dynamic 3D brain volumes with high spatial resolution. The voxel-wise deformation was derived from jacobian maps over the cardiac cycle. We found that greater deformation in the brain occurs in basal ganglia and brain stem, and then attenuates toward the white matter and brain cortex during the cardiac cycle.

Melissa Haskell^1,2, Stephen Cauley^1,3, and Lawrence Wald^1,3,4
1Athinoula A. Martinos Center for Biomedical Imaging, MGH/HST, Charlestown, MA, United States, ²Graduate Program in Biophysics, Harvard University, Cambridge, MA, United States, ³Harvard Medical School, Boston, MA, United States, ⁴Harvard-MIT Division of Health Sciences and Technology, MIT, Cambridge, MA, United States

We approach the reconstruction of artifact-free images from an object undergoing unknown rigid-body transformations using a joint optimization of the final uncorrupted image and motion parameters. To characterize motion, the joint optimization must estimate 6 additional parameters for each shot in the image acquisition. We demonstrate an efficient method for reconstruction from translation-corrupted kspace data by examining iterative improvements to only a small, targeted subset of imaging voxels. The method can be enhanced by providing incomplete or noisy information from motion sensors or navigator measurements. We discuss generalizing our hybrid greedy and global step non-linear optimization to full rigid-body motion.

Rüdiger Stirnberg¹, Daniel Brenner¹, Willem Huijbers¹, Tobias Kober^2,3,4, and Tony Stöcker^1,5
1German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany, ²Advanced Clinical Imaging Technology, Siemens Healthcare, Lausanne, Switzerland, ³Department of Radiology, University Hospital Lausanne (CHUV), Lausanne, Switzerland, ⁴Department of Electrical Engineering (LTS5), Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, ⁵Department of Physics and Astronomy, University of Bonn, Bonn, Germany

Accurate and precise head motion tracking has been shown to be feasible using multi-channel free-induction-decay (FID) signals, where positional information is supported by the spatial distribution of the receive coils. Until now, this required subject-specific calibration using simultaneously acquired FID signals and reference motion parameters, e.g. from an external device, while the subject performs controlled motion. In this study, we demonstrate successful calibration of FID navigators using motion parameters extracted from a resting-state fMRI scan without intentional motion. Additionally, extension of the calibration by principal component analysis of the FID data is shown to increase motion prediction accuracy and precision. 

Niklas Wehkamp¹, Benjamin Richard Knowles ¹, Patrick Hucker¹, and Maxim Zaitsev^1
1Department of Radiology - Medical Physics, University Medical Center Freiburg, Freiburg, Germany

Marker fixation remains an unresolved issue in Prospective Motion Correction (PMC) using optical tracking. The most common and simple approach to track motion of the skull is using markers adhered to the face. However, markers applied in this fashion can report erroneous positions due to facial gestures of the subject during the MR examination. The presented approach using multiple markers is a patient friendly solution, offering robust position data for PMC in the presence of facial gestures. The presented approach is a promising solution to stabilize prospective motion tracking and thus to significantly reduce costs for MR imaging facilities.

Daniel V Litwiller¹, Erik Tryggestad², Kiaran McGee³, Yuji Iwadate⁴, Lloyd Estkowski⁵, and Ersin Bayram^6
1Global MR Applications & Workflow, GE Healthcare, New York, NY, United States, ²Department of Radiation Oncology, Mayo Clinic, Rochester, MN, United States, ³Department of Radiology, Mayo Clinic, Rochester, MN, United States, ⁴Global MR Applications & Workflow, GE Healthcare, Hino, Tokyo, Japan, ⁵Global MR Applications & Workflow, GE Healthcare, Menlo Park, CA, United States, ⁶Global MR Applications & Workflow, GE Healthcare, Houston, TX, United States

One of the motivations for 4D MRI is the need to characterize patient respiratory motion in the context of radiation therapy treatment planning (RTP).  Here, we compare two approaches for generating 4D MRI data using dynamic, navigator-based acquisitions with retrospective respiratory compensation, including single-shot balanced-SSFP and single-shot fast spin echo with variable refocusing flip angle (vrfSSFSE).  The results presented suggest that both sequences offer a straightforward approach to generating 4D MRI data for MR-guided RTP.

Anja Jäger^1,2, Thomas Beck², and Andreas Maier^1
1Pattern Recognition Lab, Department of Computer Science, Friedrich-Alexander- Universität Erlangen-Nürnberg, Erlangen, Germany, ²Siemens Healthcare, MR Application Development, Erlangen, Germany

A method for detection of patient motion based on sub-volumes is presented. Current methods for image-based motion detection are limited because rigid motion parameters can only be detected for full volumes. This limits the potential of navigator acceleration and causes undesirable effects due to respiratory motion in some applications. Our novel approach extends the rigid-body-motion model by detection based on a subset of slices relative to a fully sampled reference volume. It is validated with phantom and in-vivo data and allows for both considerable acceleration of navigator scans and prospective correction of head motion in fMRI applications.

Rong Guo¹, Yoann Petibon², Yixin Ma¹, Kui Ying¹, and Jinsong Ouyang^2
1Engineering Physics, Tsinghua University, Beijing, China, People's Republic of, ²Center for Advanced Radiological Sciences, Massachusetts General Hospital, Boston, MA, United States

Bias may be introduced in the estimation of the PET myocardial kinetic parameters by both cardiac and respiratory motion. Simultaneous PET-MR makes it possible to perform MR-based PET motion correction. We have investigated the performance of MR-based motion correction on the estimation of myocardial PET kinetic parametermsat 

Vanessa Stahl¹, Martin T. Freitag², Armin M. Nagel^1,3, Ralf O. Floca⁴, Moritz C. Berger ¹, Jan P. Karch⁵, Peter Bachert¹, Mark E. Ladd¹, and Florian Maier^1
1Medical Physics in Radiology, German Cancer Research Center, Heidelberg, Germany, ²Department of Radiology, German Cancer Research Center, Heidelberg, Germany, ³Department of Diagnostic and Interventional Radiology, University Medical Center Ulm, Ulm, Germany, ⁴Medical and Biological Informatics, German Cancer Research Center, Heidelberg, Germany, ⁵Institue of Physics, Johannes Gutenberg University Mainz, Mainz, Germany

Human brown adipose tissue (BAT) is mostly found in cervical and mediastinal anatomic sites, making MR-imaging challenging because of susceptibility to breathing motion artifacts. Image acquisition under breath-hold requires data registration, especially for long measurement times. Two deformable registration algorithms (Fast Symmetric Forces Demons (FSF), Level Set Motion (LSM)) were evaluated regarding their suitability for compensation of deviations in breath-hold positions. Data processing was based on a volunteer study using distinct anatomical landmarks placed by an experienced radiologist. Landmark positions were evaluated after transformation, showing that FSF is more suitable for registration of thoracic data allowing for human BAT assessment.

Pieter F Buur¹, Wietske van der Zwaag¹, José P Marques², and Daniel Gallichan^3
1Spinoza Centre for Neuroimaging, Amsterdam, Netherlands, ²Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, Nijmegen, Netherlands, ³Centre d'Imagerie BioMédicale (CIBM), EPFL Lausanne, Lausanne, Switzerland

Motion correction using interleaved fat navigators is a promising approach for high-resolution brain imaging at 7 Tesla. We have implemented a 3D-EPI fat navigator to reduce acquisition time and thereby minimize overhead in sequences with little or no dead time. The efficacy of motion-induced artefact removal using the fat navigators is demonstrated for 0.6 mm isotropic inversion-prepared (MPRAGE) and 0.5 mm isotropic non-prepared 3D TFE (GRE) protocols.

Paul Chang^1,2, Sahar Nassirpour^1,2, Ariane Fillmer^3,4, and Anke Henning^1,3
1Max Planck Institute for Biological Cybernetics, Tuebingen, Germany, ²IMPRS for Cognitive and Systems Neuroscience, Eberhard Karls University of Tuebingen, Tuebingen, Germany, ³Institute for Biomedical Engineering, UZH and ETH Zurich, Zurich, Switzerland, ⁴Physikalisch-Technische Bundesanstalt, Berlin, Germany

The task of mapping B₀ fields to characterise shim fields can be challenging since shim fields generate a highly inhomogeneous field that may be difficult to capture. Furthermore this results in geometric distortion of the B₀ map which affects the characterisation of the shim field.

Geometric distortion correction was investigated using a gridded phantom and compared to the effect of using a high bandwidth on the read-out gradient. It was found that using a high bandwidth was more effective in reducing the distortion and that correcting the distortion using a phantom grid was not sufficient.

Takahiko Kaneda¹, Kazuya Oshinomi¹, Naoki Ohno², Toshiaki Miyati², and Toru Yamamoto^3
1Graduate School of Health Sciences, Hokkaido University, Sapporo, Japan, ²Division of Health sciences, Graduate School of Medical Sciences, Kanazawa University, Kanazawa, Japan, ³Faculty of Health Sciences, Hokkaido University, Sapporo, Japan

Evidence that the magnetic field distortion of an artificial knee joint and an Elgiloy rod in ppm unit does not change at 0.4-T and 3.0-T MRI was demonstrated. The susceptibility of Elgiloy and Ti alloy rods was derived from the magnetic distortion maps and the value of Ti alloy rod especially agreed with the known susceptibility of its material. The obtained value of the magnetic field distortion is quantitatively reliable.

Michael J van Rijssel¹, Josien P W Pluim¹, Peter R Luijten¹, Alexander J Raaijmakers¹, and Dennis W J Klomp^1
1Center for Image Sciences, UMC Utrecht, Utrecht, Netherlands

Quantitative DCE-MRI requires reliable B₁⁺ information. This study presents a simulation-based fast B₁⁺ estimation method for DCE breast imaging at 7T. Numerical FDTD simulations were conducted to assess the inter-subject differences in B₁⁺ for four volunteers using segmented breast images for the simulation model. Inter-subject differences are shown to be comparable to the accuracy of popular B₁⁺ mapping methods, justifying the use of one generic B₁⁺ distribution for B₁⁺ estimation (coil template). This template was created by averaging the simulated B₁⁺ distributions over the four volunteers. We demonstrate the feasibility of this method in three in-vivo cases.

Nadia Karina Paschke¹, Andreas Neubauer¹, and Lothar R Schad^1
1Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany

Sodium MRI suffers from low signal-to-noise ratio, which can be compensated by applying surface coils fitting the geometry of interest. Inhomogeneous coil profiles hinder absolute quantification of in vivo tissue sodium concentration, which is crucial for clinical assessment of pathological changes. Adequate corrections of intensity inhomogeneities of reception radiofrequency fields are essential and most standard proton imaging correction methods require manual thresholding. We present a novel and automatic correction approach by postprocessing images with Ensemble Empirical Mode Decomposition without additional scan time. It reduces signal variations by 39%. This is shown in phantoms and in vivo.

Andreas Lesch¹, Matthias Schlögl¹, Martin Holler², and Rudolf Stollberger^1,3
1Institute of Medical Engineering, Graz University of Technology, Graz, Austria, ²Institute of Mathematics and Scientific Computing, University of Graz, Graz, Austria, ³BioTechMed Graz, Graz, Austria

In this work we describe a novel method, which is able to reconstruct B₁₊-maps from highly under-sampled Bloch-Siegert data. This method is based on variational methods and a problem specific regularization approach.  We show its capability to achieve successful reconstructions from more than 100times under-sampled 3D-data in the human brain with a mean error below 1%. The results are compared to a fully-sampled reference and a conventional low resolution reconstruction for different under-sampling factors.

Gary George McGinley^1,2, Atle Bjørnerud^3,4, and Øystein Bech Gadmar^3
1Institute for Experimental Medical Research, Oslo University Hospital, Oslo, Norway, ²KG Jebsen Cardiac Research Center and Center for Heart Failure Research, University of Oslo, Oslo, Norway, ³The Intervention Centre, Oslo University Hospital, Oslo, Norway, ⁴Department of Physics, University of Oslo, Oslo, Norway

This study aims to compare the effectiveness of three reverse-gradient method susceptibility artefact correction tools (EPIC, TOPUP, and HySCO) in the correction of spin-echo (SE) and gradient-echo (GE) EPI images of the brain, and to measure the effect of pixel bandwidth, SENSE factor and slice thickness on artefact correction. This was achieved by co-registering the artefacted and corrected images to an anatomical scan and measuring the normalised mutual information (NMI). It was found that EPIC correction resulted in the largest gains in NMI and that more mutual information was recovered at lower pixel bandwidths after EPIC correction.

Taejoon Eo^1
1Yonsei University, Seoul, Korea, Democratic People's Republic of

We propose a robust phase unwrapping method for low-SNR multi-echo MR images based on complex signal modeling. This method is superior to conventional phase unwrapping methods and provides high-quality unwrapped phase images without any spatial artifacts caused by high noise.

Ying Chen¹, Song Chen¹, Hui Liu², and Jianhui Zhong^1,3
1Center for Brain Imaging Science and Technology, Zhejiang University, Hangzhou, China, People's Republic of, ²MR Collaboration Northeast Asia, Siemens Healthcare, Shanghai, China, People's Republic of,³Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University, Hangzhou, China, People's Republic of

Single-shot GE-EPI is widely used in fMRI. However, it is susceptible to field inhomogeneity induced geometric distortions, therefore retrospectively unwarping of the single-shot GE-EPI data is important. A commonly used unwarping technique is based on the field map of the image and it would be desirable to acquire the field map at each time point of a dynamic fMRI measurement series. The aim of this abstract is to qualitatively and quantitatively compare the performance of three reference-free unwarping methods on human brain imaging data. Experimental results demonstrate that the field map obtained from measuring the k-space shifts of each voxel can provide more reliable unwarped images.

Parnian Zarghamravanbakhsh¹, John M Pauly¹, and Greig Scott^1
1Electrical Engineering, Stanford University, Stanford, CA, United States

Accurate knowledge of magnetic field distribution is necessary for RF shimming and calibration of parallel transmit systems. The incident field distribution depends on relative location of transmit array to sample, also magnitude and phase of coil current(thereby, magnetic field) vary with different sample loading. The RF maps of each coil can be estimated by localizing the transmit array in the image space and circulating RF currents in each coil. In this study, parallel transmit system RF shimming pulses are designed by using transmit coils locator information and coil current measurement without performing experimental B₁ mapping.

Taisuke Harada^1,2, Kohsuke Kudo¹, Ikuko Uwano³, Fumio Yamashita³, Hiroyuki Kameda^1,3, Tsuyoshi Matuda⁴, Makoto Sasaki³, and Hiroki Shirato^2
1Department of Diagnostic and Interventional Radiology, Hokkaido University Hospital, Sapporo, Japan, ²Department of Radiation Medicine, Hokkaido University Graduate School of Medicine, Sapporo, Japan,³Division of Ultrahigh Field MRI, Institute for Biomedical Sciences, Iwate Medical University, Yahaba, Japan, ⁴MR Applications and Workflow, GE Healthcare, Tokyo, Japan

The aim of our study was to compare the reproducibility of those maps and GRASS images of brain scanned with pTx at 7T between free-breathing (FB) and breath-holding (BH) during the calibration scan. Nine healthy volunteers were scanned by 7T MRI with RF shimming, RF design of pTx, and quadrature transmission (qTx). The reproducibility of B0 and B1+ were better in BH than FB, and the same results was seen in GRASS images. The intensity homogeneity was not different between qTx and RF shimming however was better in RF design than qTx. These results might facilitate the development of pTx.

Jürgen Finsterbusch^1
1Systems Neuroscience, University Medical Center Hamburg-Eppendorf, Hamburg, Germany

T2*-weighted acquisitions used for functional neuroimaging can suffer from through-slice dephasing. Additional, so-called z-shim gradient pulses can be applied in the slice direction to minimize related signal losses. Thereby, a single, slice-specific gradient moment may be sufficient for small target regions like the spinal cord. To combine this z-shimming approach with multi-band acceleration, the individual bands must provide different dephasing moments in the slice direction. This can be realized with appropriate temporal shifts of the individual envelopes as is demonstrated for conventional slice-selective RF excitations and 2D-selective RF excitations for inner-field-of-view imaging.

Rebecca Emily Feldman¹, Hadrien A Dyvorne¹, Rafael O'Halloran¹, and Priti Balchandani^1
1Translational and Molecular Imaging Institute, Icahn School of Medicine at Mount Sinai, New York, NY, United States

The higher signal-to-noise-ratio offered at 7T, has been shown to increase the resolution of diffusion MRI as well as the precision and directional certainty of diffusion-based parameters. Two major drawbacks of 7T dMRI include lengthy acquisitions and signal loss due to B₁-inhomogeneity. SMS methods reduce the duration of the acquisition, the refocusing pulses typically used in dMRI are particularly sensitive to B1 non-uniformities leading to a loss in signal, or even complete signal dropout in parts of the image. We have created a dMRI sequence with SEAMS PINS and an EPI readout that provides improved immunity to B₁-inhomogeneity. 

Samy Abo Seada¹, Joseph V Hajnal¹, and Shaihan J Malik^1
1Division of Imaging Sciences and Biomedical Engineering, King's College London, London, United Kingdom

Simultaneous multi-slice imaging can accelerate image acquisition for commonly used diffusion and functional MRI sequences. The design of multiband pulses can be problematic due to their high peak amplitude. Another issue is that the necessary rapid phase and amplitude modulation can be problematic for some current MRI RF systems to reproduce. Phase related issues can be avoided by designing purely amplitude modulated waveforms. We describe how three current multiband pulse design techniques (phase optimisation, time shifting and root-flipping) can be modified to produce purely amplitude modulated pulses and find that the relative peak increase is only about 20-25%.

Ying Chen¹, Song Chen¹, Zhong Chen², and Jianhui Zhong^1,3
1Center for Brain Imaging Science and Technology, Zhejiang University, Hangzhou, China, People's Republic of, ²Department of Electronic Science, Xiamen University, Xiamen, China, People's Republic of, ³Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University, Hangzhou, China, People's Republic of

In-plane reduced field-of-view excitation based on two-dimensional radio-frequency pulse (2DRF) has been widely used in many applications. The EPI-style gradient waveform is commonly used in 2DRF implementation. However, at high field, the off-resonance effects during excitation would result in distortions of the profiles obtained. This work is to investigate the feasibility to achieve in-plane selective excitation of anatomically pre-defined regions using SPEN-2DRF pulse under different shim conditions. Experimental results show that the proposed method can produce profiles with significantly improved robustness to distortions at high field than the Fourier-based 2DRF pulse.

Hong Shang¹, Hai Luo², Xia Liu², Gaojie Zhu², and Leping Zha^2
1Bioengineering, UC Berkeley - UCSF, Berkeley/San Francisco, CA, United States, ²AllTech Medical Systems, Chengdu, China, People's Republic of

Two classes of previously proposed nonlinear phase RF pulses, the quadratic-phase pulse and root-flip optimized pulse, are compared in terms of selectivity, peak B₁ value, pulse energy, and sensitivity to B₁variations, when applied for spatial outer volume suppression. Root-flip pulses have lower peak B₁ and energy given the same transition width and pulse duration, or sharper transition given the same peak B₁, while quadratic-phase pulses have less sensitivity to B₁ variations that maintains profile shape with B₁ deviations, and thus less prone to residual saturation band magnetization. This work provides insights to pulse designers in regards to nonlinear phase pulse design and application.

David Y. Zeng¹, Jieying Luo¹, Dwight G. Nishimura¹, and Adam B. Kerr^1
1Electrical Engineering, Stanford University, Stanford, CA, United States

A B₁-insensitive T₂-weighted preparation sequence with integrated fat saturation and outer volume suppression for localized cardiac imaging is proposed. The sequence is composed of a BIR-4 90° tip-down pulse, two spectral-spatial adiabatic refocusing pulses and a BIR-4 -90° tip-up pulse. Outer volume suppression is achieved by the spatial selectivity of the first refocusing pulse in x and spatial selectivity of the second refocusing pulse in y. Fat suppression is achieved by spectral selectivity of the refocusing pulses. Numerical simulation and phantom experiments verify the performance of the sequence.

Seohee So¹, HyunWook Park¹, Dongchan Kim¹, Hyunseok Seo¹, JaeJin Cho¹, Young Woo Park¹, and Kinam Kwon^1
1Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, Korea, Republic of

  Slice profile used for magnetic resonance (MR) imaging has transition region between passband and stopband. Sharper transition performs better slice selection. This abstract proposes method to design short-duration RF pulse without increasing transition region width. Additional phase is merged into a transition region of slice profile. Two RF pulses having different phases are used alternately. The proposed algorithm produces about 30% reduction of RF pulse duration without transition increase.

Lucas Soustelle¹, Paulo Loureiro de Sousa¹, Sascha Koehler², Chrystelle Po¹, François Rousseau³, and Jean-Paul Armspach^1
1Université de Strasbourg, CNRS, ICube, FMTS, Strasbourg, France, ²Bruker BioSpin MRI, Ettlingen, Germany, ³Institut Mines Télécom, Télécome Bretagne, INSERM LaTIM, Brest, France

The Variable-rate Selective Excitation (VERSE) approach allows to achieve very short echo time in 2D-UTE sequences when applied on a selective half-pulse and its paired slice selection gradient. Unfortunately, the latter may suffer from non-linearities and eddy current effects, all the more important on preclinical scanners equipped with strong gradient systems. 
An efficient method was implemented on a 7T preclinical scanner to measure the real slice selection gradient profile. A reshaping of the corresponding pulse was made, improving the slice selectivity.

 

Ethan M Johnson¹, Kim Butts Pauly², Pejman Ghanouni², and John M Pauly^1
1Electrical Engineering, Stanford University, Stanford, CA, United States, ²Radiology, Stanford University, Stanford, CA, United States

A method for sensitising 3D UTE sequences to the short-$$$T_2$$$ range of cortical bone using scaled RF hard pulses has been previously demonstrated for creating CT-like contrast in MR imaging.  However whole-volume excitation and encoding is not practical in all contexts.  Here, an adaptation for slice-selective half-pulses is presented that enables 2D image encoding for MR-simulated-CT images.

Yi-Cheng Hsu¹, Ying-Hua Chu¹, and Fa-Hsuan Lin^1
1Institute of Biomedical Engineering, National Taiwan University, Taipei, Taiwan

A new simultaneous multi-slice excitation method was proposed to reduce RF power delivery by only concerning slice profiles within the imaging object. Compared to MultiPINS, our approach used only 79% of the RF energy to the same result (slice thickness = 3mm, MB factor = 5, 4 bandwidth time product, excitation duration 6380 $$$\mu s$$$). This excitation method was experimentally demonstrated in spin-echo EPI with blipped CAIPI acquisition.

Christoph Stefan Aigner^1,2, Christian Clason³, Armin Rund⁴, and Rudolf Stollberger^1,2
1Institute of Medical Engineering, Graz University of Technology, Graz, Austria, ²BioTechMed Graz, Graz, Austria, ³Faculty of Mathematics, University of Duisburg-Essen, Essen, Germany, ⁴Institute for Mathematics and Scientific Computing, University of Graz, Graz, Austria

We demonstrate the joint optimization of RF and slice selective gradient shapes with hard constraints such as peak B1 of the pulse and peak slew rate of the gradient via a flexible approach based on optimal control of the full time-dependent Bloch equation and a novel semi-smooth Newton method. The presented approach allows optimization on a fine spatial and temporal grid while enforcing physical and technical limitations on the control variables. The results are validated on a 3T scanner, demonstrating the practical realizability of the presented approach even for short RF pulses.

David G Norris^1,2 and Jenni Schulz^1
1Donders Institute for Brain Cognition and Behaviour, Radboud University Nijmegen, Nijmegen, Netherlands, ²Erwin L Hahn Institute, University Duisburg Essen, Essen, Germany

This abstract describes the use of pulses that have a quadratic phase profile in the spatial dimension for simultaneous multi-slice imaging. The quadratic profile reduces the peak voltage needed by an amount dependent on the number of simultaneously excited slices. The pulses have to be used as an excitation-refocusing pair, with the refocusing pulse having half the phase gradient. The echoes from each slice are simultaneously refocused and there are no additional constraints on the pulse duration.

Ronald Mooiweer¹, Shaihan J Malik², Joseph V Hajnal², Nico van den Berg¹, Peter R Luijten¹, and Hans Hoogduin^1
1UMC Utrecht, Utrecht, Netherlands, ²Division of Imaging Sciences and Biomedical Engineering, King’s College London, London, United Kingdom

In this work the design of SPINS excitation pulses has been expanded for use in TSE sequences and was compared to dynamic RF shimming using DSC in a standard T2w TSE sequence. We have demonstrated homogeneous 90 degree excitation, but in itself this was not sufficient to make TSE images uniform. Manipulating the refocusing pulses (using DSC) remains a necessity. 

Eric Kenneth Gibbons¹, John Mark Pauly², and Adam Bruce Kerr^2
1Department of Bioengineering, Stanford University, Stanford, CA, United States, ²Department of Electrical Engineering, Stanford University, Stanford, CA, United States

SS-FSE is a robust method for fast image acquisition in areas where there is significant B0 inhomogeneity.  Recent efforts have led to expand the capabilities beyond traditional constraints of SS-FSE meeting the CPMG condition.  In this work, we examine the effects of various RF pulse types on the stability of the signal using a quadratic phase modulation as well as propose using a novel DIVERSE pulse.

Frank Ong¹ and Michael Lustig^1
1Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA, United States

We propose a generalized Shinnar-Le-Roux transform that maps $$$T_1$$$, $$$T_2$$$ and frequency selective pulses to multi-dimensional polynomials. We show that the polynomial mapping is one-to-one and hence designing these RF pulses reduces to multi-dimensional polynomial design. We describe a convex approach to the multi-dimensional polynomial design and show preliminary $$$T_2$$$ and frequency selective pulses.

Dietmar Cordes^1,2, Zhengshi Yang¹, Xiaowei Zhuang¹, Karthik Sreenivasan¹, and Le Hanh Hua^1
1Cleveland Clinic Lou Ruvo Center for Brain Health, Las Vegas, NV, United States, ²Department of Psychology and Neuroscience, University of Colorado, Boulder, CO, United States

In this study, a new algorithm to better model the receiver coil sensitivities with the purpose of obtaining unbiased proton density maps is proposed. Using optimized orthonormal basis functions for the modeling produces an accurate fit of potential inhomogeneities of the signal due to receiver coil bias. The obtained final image of the proton density has low variance, suitable for quantitative diagnostic information of brain tissue. Results are shown for nine MS patients and one control subject.  

Anna Scherman Rydhög¹, André Ahlgren¹, Freddy Ståhlberg^1,2,3, Ronnie Wirestam¹, and Linda Knutsson^1
1Department of Medical Radiation Physics, Lund University, Lund, Sweden, ²Department of Diagnostic Radiology, Lund University, Lund, Sweden, ³Lund Bioimaging Center, Lund University, Lund, Sweden

Intravoxel Incoherent Motion (IVIM) is a method for quantification of perfusion parameters, such as the perfusion fraction Fb. Unfortunately, CSF partial volume effects are often seen in the estimated blood compartment. This work introduces a novel version of the IVIM model, containing three compartments (tissue, CSF and blood), where multi-TE and multi-TI data are incorporated to yield a direct relaxation estimate. Using this relaxation-compensated model, results were obtained from in vivo measurements in a volunteer. Compared to a non-relaxation-compensated model, the three-compartment model with relaxation-compensated data reduced the CSF contamination.

Pierre Besson^1,2,3, Arnaud Le Troter^1,2, Julien Sein^1,2, Gilles Brun^1,2, Maxime Guye^1,2, and Jean-Philippe Ranjeva^1,2
1Aix-Marseille Université, CNRS, Centre de Résonance Magnétique Biologique et Médicale (CRMBM) UMR 7339, Marseilles, France, ²APHM, Timone Hospital, Pôle d’Imagerie, Centre d’Exploration Métabolique par Résonance Magnétique (CEMEREM), Marseilles, France, ³Siemens Healthcare, St Denis, France

UHF 7T MR scanners offers the possibility to acquire very high resolution in-vivo images, providing a new insight into human brain structural characterization. Nevertheless, in order to obtain highly contrasted and highly spatially resolved atlas, and to compensate for the drop in SNR related to reduction of the voxel size, averaging data among several subjects is needed. We present in this abstract an automatic pipeline that generates a whole brain high-resolution T1-weighted template (called 7TAMIbrainT1w_30) built from MP2RAGE acquisitions obtained in 30 healthy controls at 7T.

Ruth Oliver^1,2, Linda Ly^1,2, Chenyu Wang^1,2, Heidi Beadnall², Ilaria Boscolo Galazzo^3,4, Michael Chappell^5,6, Xavier Golay⁷, Enrico De Vita⁷, David Thomas⁷, and Michael Barnett^1,2
1Sydney Neuroimaging Analysis Centre, Sydney, Australia, ²University of Sydney, Sydney, Australia, ³Institute of Nuclear Medicine, University College London, London, United Kingdom, ⁴Department of Neuroradiology, University Hospital Verona, Verona, Italy, ⁵Institute of Biomedical Engineering, University of Oxford, Oxford, United Kingdom, ⁶FMRIB Centre, University of Oxford, Oxford, United Kingdom, ⁷Institute of Neurology, University College London, London, United Kingdom

ASL is a low resolution imaging modality that suffers from the partial volume effect, leading to an underestimation of GM perfusion. This effect has two principle causes; blurring from the point spread function in the slice direction, and inadequate resolution due to the need for large voxels to achieve sufficient SNR. Both may act as confounders for measurement of GM CBF abnormalities. Decreased GM perfusion could reflect neuronal loss or metabolic dysfunction; PV correction allows a decoupling of structure and function. We present the first application of a complete PV correction solution for ASL to a cohort of MS patients.

Shanlin Qin¹, Fawang Liu¹, Ian William Turner^1,2, Qiang Yu³, Qianqian Yang¹, and Viktor Vegh^3
1School of Mathematical Sciences, Queensland University of Technology, Brisbane, Australia, ²ARC Centre of Excellence for Mathematical and Statistical Frontiers, Melbourne, Australia, ³Centre for Advanced Imaging, University of Queensland, Brisbane, Australia

MRI models based on integer order calculus lack the ability to accurately map magnitude signal decay in the human brain, likely due to magnetic susceptibility and microstructure variations in tissues. We applied fractional calculus to the Bloch equation with the aim of developing a model capable of matching experimental findings. Solution of the time-fractional Bloch equation resulted in a new five parameter model. We analysed model parameters in nine brain regions using multiple echo gradient recalled echo MRI data from five participants. Time-fractional model parameters may provide new ways of studying microstructure and susceptibility induced changes in the human brain.

Huanjie Li¹, Lisa D. Nickerson², Yang Fan³, Thomas E. Nichols⁴, and Jia-Hong Gao^5
1Department of Biomedical Engineering, Dalian University of Technology, Dalian, China, People's Republic of, ²McLean Imaging Center, McLean Hospital/Harvard Medical School, Belmont, MA, United States, ³GE Healthcare, MR Research China, Beijing, China, People's Republic of, ⁴Department of Statistics and Warwick Manufacturing Group, University of Warwick, Coventry, United Kingdom, ⁵Center for MRI Research, Peking University, Beijing, China, People's Republic of

Threshold-free cluster enhancement (TFCE) outperforms the cluster-size test (CST) based on random field theory and our recent papers provide two voxelation-corrected CST (v-CST and vn-CST) which also show the clear advantage over other CST as well. However, it’s not clear which one shows better performance for MRI data analysis. This work provides a very careful, fair and thorough evaluation of the powerful statistical methods,  which may be particularly appealing for group-level MRI data analysis.

Alexandre Bizeau^1,2, Guillaume Gilbert³, Minh Tung Huynh⁴, Michaël Bernier^1,2, Christian Bocti⁵, Maxime Descoteaux^2,6, and Kevin Whittingstall^1,2,4
1Department of Radiation Sciences and Biomedical imagery, Université de Sherbrooke, Sherbrooke, QC, Canada, ²Centre d’Imagerie Moléculaire de Sherbrooke (CIMS), Centre de Recherche CHUS, Sherbrooke, QC, Canada, ³MR Clinical Science, Philips Healthcare, Markham, ON, Canada, ⁴Department of Diagnostic Radiology, Université de Sherbrooke, Sherbrooke, QC, Canada, ⁵Department of Medecine, Université de Sherbrooke, Sherbrooke, QC, Canada, ⁶Department of Computer Science, Université de Sherbrooke, Sherbrooke, QC, Canada

When undergoing stimulation, neurons need to be supplied with oxygen and glucose. This demand then induces vasodilation generated by the astrocytes which act on the muscles of the arteries of the human brain. Using time-of-flight magnetic resonance angiography acquisitions, we extracted the apparent diameter of arterial vessels. We then compared diameter with and without visual stimulation and demonstrated that smaller vessels dilate proportionally more than larger ones in the posterior cerebral arteries. Using this method, the investigation of the coupling between neural activity and regional cerebral vasodilation, also called functional hyperhemia, is now possible. 

Thomas Küstner^1,2, Martin Schwartz^1,2, Annika Kaupp², Petros Martirosian¹, Sergios Gatidis¹, Nina F. Schwenzer¹, Fritz Schick¹, Holger Schmidt¹, and Bin Yang^2
1University Hospital Tübingen, Tübingen, Germany, ²Institute of Signal Processing and System Theory, University of Stuttgart, Stuttgart, Germany

Acquired images are usually analyzed by a human observer (HO) according to a certain diagnostic question. Flexible algorithm parametrization and the enormous amount of data created per patient make this task time-demanding and expensive. Furthermore, definition of objective quality criterion can be very challenging, especially in the context of a missing reference image. In order to support the HO in assessing image quality, we propose a non-reference MR image quality assessment system based on a machine-learning approach with an Active Learning loop to reduce the amount of necessary labeled training data. Labeling is performed via an easy accessible website.

Mu Lin¹, Xiaozhi Cao¹, Congyu Liao¹, Xu Yan², and Jianhui Zhong^1
1Center for Brain Imaging Science and Technology, Zhejiang University, Hangzhou, China, People's Republic of, ²MR Collaboration NE Asia, Siemens Healthcare, Hangzhou, China, People's Republic of

Multi-component tissue model with priori T₁ and T₂ have been used to decompose MRF data. We propose that tissue classification can be improved when the selection uses clustering method based on cross-correlation. Our results from phantom and in vivo measurements show that the method successfully separates signal from different tissue types, allows extraction of tissue fractions, and results are more robust with image quality.

Jannis Hanspach¹, Michael G Dwyer¹, Niels P Bergsland^1,2, Xiang Feng³, Jesper Hagemeier¹, Paul Polak¹, Nicola Bertolino¹, Jürgen R Reichenbach^3,4, Robert Zivadinov^1,5, and Ferdinand Schweser^1,5
1Buffalo Neuroimaging Analysis Center, Department of Neurology, Jacobs School of Medicine and Biomedical Sciences, The State University of New York at Buffalo, Buffalo, NY, United States, ²MR Research Laboratory, IRCCS Don Gnocchi Foundation ONLUS, Milan, Italy, ³Medical Physics Group, Department of Diagnostic and Interventional Radiology, Jena University Hospital - Friedrich Schiller University Jena, Jena, Germany,⁴Michael Stifel Center for Data-driven and Simulation Science Jena, Friedrich Schiller University Jena, Jena, Germany, ⁵MRI Molecular and Translational Research Center, Jacobs School of Medicine and Biomedical Sciences, The State University of New York at Buffalo, Buffalo, NY, United States

Quantitative susceptibility mapping (QSM) is a recent in vivo magnetic resonance imaging (MRI) technique that provides quantitative information about the bulk magnetic susceptibility distribution in tissues, a promising measure for studying brain iron. A voxel-based analysis (VBA) of susceptibility maps would facilitate a better understanding of the intricate anatomical structure (e.g. sub-nuclear regions) of deep gray matter and its relation to diseases and normal aging.

In the present work, we developed and quantitatively assessed six strategies for creating a susceptibility brain template for VBA based on ANTs, representing the first step toward an understanding of sub-nuclear susceptibility changes without the need for a priori information.

Serena Monti^1,2, Pasquale Borrelli¹, Sirio Cocozza³, Sina Straubb⁴, Mark Ladd⁴, Marco Salvatore¹, Enrico Tedeschi³, and Giuseppe Palma^5
1IRCCS SDN, Naples, Italy, ²Department of Electronics, Information and Bioengineering, Politecnico di Milano, Milan, Italy, ³Department of Advanced Biomedical Sciences, University "Federico II", Naples, Italy,⁴Department of Medical Physics in Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany, ⁵Institute of Biostructure and Bioimaging, National Research Council, Naples, Italy

A new fully automated algorithm, based on structural, morphological and relaxometric information, is proposed to segment the entire brain deep venous system from MR images. The method is tested on brain datasets at different magnetic fields and its inter-scan reproducibility is also assessed. The proposed segmentation algorithm shows good accuracy and reproducibility, outperforming previous methods and becoming a promising candidate for the characterization of venous tree topology. 

Yash Agarwal¹, Philippe Young¹, Ross Cotton¹, Chris Pearce², Siddiq Qidwai³, Amit Bagchi³, and Nithyanand Kota^3
1Simpleware Ltd., Exeter, United Kingdom, ²Atkins, Epsom, United Kingdom, ³U.S. Naval Research Laboratory, Washington, DC, United States

Image-based model generation methods demonstrate the value of creating realistic human head models based on high-resolution MRI data. Head models created by the U.S. Naval Research Laboratory and Simpleware (Exeter, UK) are being used to study head impact and traumatic brain injury; this offers a solution to the problem of limited experimental testing. Results from the modelling methodology and simulation demonstrate a good level of accuracy when compared to experimental benchmarks. The methodology and models have been extended for use in areas such as examining head impact in sports including American football, rugby and cricket.

Farid Badar¹, Ji Hyun Lee¹, and Yang Xia^1
1Department of Physics and Center for Biomedical Research, Oakland University, Rochester, MI, United States

The consequences of two different zone-division methods in MRI T2 of articular cartilage were studied, using an animal model of early osteoarthritis (OA). By dividing the cartilage thickness unequally, significant improvement in OA detection can be achieved – both in the deeper cartilage as well as between the contralateral and normal tissue. This improved detection may become important in the clinical diagnostics of early OA.

Anagha Deshmane¹, Debra McGivney², Chaitra Badve³, Alice Yu⁴, Yun Jiang¹, Dan Ma², and Mark Griswold^1,2
1Biomedical Engineering, Case Western Reserve University, Cleveland, OH, United States, ²Radiology, Case Western Reserve University, Cleveland, OH, United States, ³Radiology, University Hospitals of Cleveland, Cleveland, OH, United States, ⁴School of Medicine, Case Western Reserve University, Cleveland, OH, United States

Synthetic weighted images from quantitative parameter maps suffer from partial volume artifacts which can distort contrast.  In this work, Partial Volume MR Fingerprinting is applied to estimate and remove signal due to cerebrospinal fluid (CSF) in the brain, allowing for improved contrast in synthetic FLAIR images generated from MRF relaxation time maps.  

Mohammed Goryawala¹, Bhaswati Roy², Rakesh K Gupta², and Andrew A Maudsley^1
1Department of Radiology, University of Miami, Miami, FL, United States, ²Department of Radiology, Fortis Memorial Research Institute, Gurgaon, India

Calculated differences between two images of differing T1 or T2 contrasts, or subtraction images, have been presented as a way to improve image contrast for imaging of brain tumors. In this study the performance of subtraction images for differentiation of tumor, edema, and normal appearing white matter (NAWM) is compared to traditionally acquired anatomic MRIs, diffusion tensor imaging (DTI), perfusion weighted imaging (PWI) and MR spectroscopy imaging (MRSI). Results showed a significant increase in contrast for differentiating between enhancing tumor and edematous regions from NAWM using the ΔT1 map and ΔT2 map, respectively, as compared to other parametric maps. 

Limin Li¹ and Alice M Wyrwicz^1,2
1Center for Basic MR Research, Northshore University Healthsystem, Evanston, IL, United States, ²Department of Biomedical Engineering, Northwestern University, Evanston, IL, United States

The processing rate for real-time multi-slice image reconstruction on an FPGA can be improved significantly by taking advantage of its parallel processing capability. In particular, multiple 2D FFT processors can be embedded into a single FPGA and run simultaneously. In this abstract, we report a new design of a 2D FFT processor with significant reduced usage of hardware resource. Test results show that an important type of resource, DSP48 slice, can be reduced by up to 50% without degrading processing performance, which implies that more 2D FFT cores can be installed into a single FPGA with a given size.

Takeshi Hara¹, Yuta Takeda¹, Tetsuro Katafuchi², Taiki Nozaki³, Masaki Matsusako³, and Hiroshi Fujita^1
1Intelligent Image Information, Gifu University Graduate School of Medicine, Gifu, Japan, ²Health Science, Gifu University of Medical Science, Seki, Japan, ³Radiology, St. Luke's International Hospital, Tokyo, Japan

Features in Parkinson's disease (PD) are a degeneration and loss of the dopamine neurons in striatum. 123I-FP-CIT can visualize the distribution by binding to the dopamine neurons. The radioactivated medicine is used for diagnosis of PD and Dementia with Lewy Bodies (DLB). The material can visualize activities in corpus striatum on SPECT images, but the location of the corpus striatum on SPECT images are often lost because of the low uptake. To realize a quantitative image analysis for the SPECT images, image registration technique to determine the region of corpus striatum on SPECT images are required to measure precise uptakes. In this study, we proposed an image fusion technique for SPECT and MR images by intervening CT image taken by SPECT/CT. We employed 30 cases of SPECT/CT and MR cases for the evaluation. 25 of 30 cases were registered correctly with registration errors less than 5mm. These results enable to measure precise uptake on SPECT images based on the segmentation results on MR images.

Mustapha Bouhrara¹ and Richard G. Spencer^1
1NIA, NIH, Baltimore, MD, United States

Formulations of the two-component spoiled gradient recalled echo (SPGR) and balanced steady-state free precession (bSSFP) models that incorporate nonzero echo time (TE) effects are presented in the context of mcDESPOT and compared with the conventionally used SPGR and bSSFP models which ignore nonzero TEs. Relative errors in derived parameter estimates from conventional mcDESPOT, omitting TE effects, are assessed using simulations over a wide range of experimental and sample parameters. The neglect of nonzero TE leads to an overestimate of the SPGR and an underestimate of the bSSFP signals. These effects introduce large errors in parameter estimates derived from conventional mcDESPOT.

Changqing Wang^1,2,3, Xinyuan Zhang², Yanying Ma⁴, Xiaoyun Liu¹, Diego Hernando³, Scott B. Reeder^3,5,6,7,8, Wufan Chen^1,2, and Yanqiu Feng^2
1School of Automation Engineering, University of Electronic Science and Technology of China, Chengdu, China, People's Republic of, ²School of Biomedical Engineering and Guangdong Provincial Key Laboratory of Medical Image Processing, Southern Medical University, Guangzhou, China, People's Republic of, ³Radiology, University of Wisconsin-Madison, Madison, WI, United States, ⁴School of Mathematical Sciences, University of Electronic Science and Technology of China, Chengdu, China, People's Republic of, ⁵Medical Physics, University of Wisconsin-Madison, Madison, WI, United States, ⁶Biomedical Engineering, University of Wisconsin-Madison, Madison, WI, United States, ⁷Medicine, University of Wisconsin-Madison, Madison, WI, United States, ⁸Emergency Medicine, University of Wisconsin-Madison, Madison, WI, United States

R2* mapping has the potential to provide rapid and accurate quantification of liver iron overload. However, conventional voxelwise liver R2* mapping methods are challenging when using echo images with low signal-noise ratio (SNR). The purpose of this work was to improve liver R2* mapping by a noise correction model incorporating weighted neighborhood information. Simulation and in vivo results demonstrate that the proposed method produces more accurate R2* maps with high spatial resolution compared to two recently proposed R2* mapping methods.

Max W.K. Law¹, Calvin M.H. Lee¹, Gladys G. Lo², Jing Yuan¹, Oilei Wong¹, Abby Y. Ding¹, and Siu Ki Yu^1
1Medical Physics and Research Department, Hong Kong Sanatorium & Hospital, Hong Kong, Hong Kong, ²Department of Diagnostic and Interventional Radiology, Hong Kong Sanatorium & Hospital, Hong Kong, Hong Kong

This abstract proposes a new algorithm that automatically segments the skull from gradient echo based magnetic resonance images to facilitate MR-based radiotherapy planning. The proposed algorithm compared the neighboring voxel intensity to capture local structural information of bone. The structural information was incorporated in a topology template which encapsulated global topology prior of skulls to achieve automatic segmentation. With the sequence-independent structural and topology priors, this method is potentially applicable to other scanning sequences. The segmented skull will be helpful for clinical applications such as cephalometry and MR-based radiotherapy planning to reduce ionizing-radiation received by patients.

Seiji Kumazawa¹, Takashi Yoshiura², Akihiro Kikuchi¹, Go Okuyama¹, Daisuke Shimao¹, and Masataka Kitama^1
1Hokkaido University of Science, Sapporo, Japan, ²Kagoshima University, Kagoshima, Japan

To correct the distortion in EPI due to field inhomogeneity, the information regarding the signal from adjacent points within each voxel is needed. The PSF approach can provide this information. Our purpose was to develop an image-based-method for estimating the PSF images in the distorted EPI image using T1WI. Our method synthesizes the distorted image to match the measured EPI image through the generation process of EPI image according to a single-shot EPI k-space trajectory and field inhomogeneity. The results demonstrate that the PSF image for each voxel in distorted EPI image can be estimated by proposed method using segmented T1WI instead of additional acquisitions for PSF measurement.

Jing Zhang¹, Bruce Bjornson², and Qing-San Xiang^3
1Applied Science Laboratory, GE Healthcare Canada, Vancouver, BC, Canada, ²Department of Pediatrics, University of British Columbia, Vancouver, BC, Canada, ³Department of Radiology, Department of Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada

Magnetization-prepared rapid gradient echo (MP-RAGE) has been widely used for T₁-weighted imaging. In order to overcome B₁ field inhomogeneity effect, the MP2RAGE sequence was introduced, with two complex images, GRE^TI1 and GRE^TI2, acquired at two inversion times TI₁ and TI₂. The MP2RAGE images are usually calculated from all the coils first and combined later into a final result. We propose an algorithm for  multi-channel MP2RAGE image combination with minimized resulting noise.

He Wang¹, Kaining Shi¹, Weibo Chen¹, and Guilong Wang^1
1Philips Healthcare, shanghai, China, People's Republic of

The study developed a methodology to improve the quality of the multi-b DWIs using the intrinsic multi-exponential pattern. It was evaluated on a healthy brain and compared with the mono-exponential model. In addition, its potential value of improving the robustness of IVIM was also evaluated. According to the results, the multi-exponential method can improve the image quality of the multi-b DWIs and may become an effective preprocessing way for the non-monoexponential models.

Ville Renvall¹ and Jonathan R. Polimeni^2
1Department of Neuroscience and Biomedical Engineering, Aalto University School of Science, Espoo, Finland, ²Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, United States

T1 mapping using multiple inversion time IR-EPI can provide a large number of different TI values in a short time, which can be utilized to characterize the relaxation time distributions within individual voxels, as an extension to multi-parametric fitting.

Katherine L. Wright¹, Peter Schmitt², Dan Ma¹, Anagha Deshmane³, Vikas Gulani¹, and Mark Griswold^1
1Radiology, Case Western Reserve University, Cleveland, OH, United States, ²Siemens Healthcare, Erlangen, Germany, ³Biomedical Engineering, Case Western Reserve University, Cleveland, OH, United States

This work proposes a method for the calculation of a single color image using quantitative T1 and T2 measurements acquired with Magnetic Resonance Fingerprinting. Quantitative MRF parameters are transformed and scaled with the goal of making normal tissues appear in grayscale and tissues with different T1 and T2 values (lesions) appear in color. 

Taejoon Eo¹, Dosik Hwang¹, and Jinseong Jang^1
1Yonsei University, Seoul, Korea, Democratic People's Republic of

The multi-echo SWV with the proposed complex signal modeling method can provide high-SNR and multi-contrast phase masks and SWV images. The multiplication number of the phase mask for SWV was increased up to 16 without image degradation even at the long TE of 49.8 ms. More detailed vein structures were visualized with higher- and multiple contrasts than the conventional single-echo GRE SWV.

Jan Scholz¹, Kaitlyn Easson², and Jason P Lerch^1,3
1Mouse Imaging Centre, Hospital for Sick Children, Toronto, ON, Canada, ²Department of Biomedical and Molecular Sciences, Queen's University, Toronto, ON, Canada, ³Department of Medical Biophysics, Department of Medical Biophysics, Toronto, ON, Canada

Most image registration algorithms do not output any information about the variance of the transformation estimates. Here we show that by perturbing input files we can recover this information without modifying the underlying algorithms. We demonstrate that local brain volume estimates can be improved by using the determinant of the average across the distribution of transformations. Our methods will improve morphological analyses, registration-based label alignment, and help find optimal registration parameters.

Martin Schwartz^1,2, Günter Steidle¹, Petros Martirosian¹, Bin Yang², and Fritz Schick^1
1Section on Experimental Radiology, Department of Radiology, University of Tuebingen, Tuebingen, Germany, ²Institute of Signal Processing and System Theory, University of Stuttgart, Stuttgart, Germany

The segmentation of signal voids, which occur in time-series of single-shot diffusion-weighted images, is important for an accelerated evaluation providing larger studies on this phenomenon. The proposed segmentation is based on a two-stage detection and segmentation approach, which utilizes a graph-based representation with random walker optimization. It was demonstrated that the presented method enables a fast and accurate segmentation of signal voids in time-series of diffusion-weighted images.

Shrushrita Sharma¹ and Yunyan Zhang^2
1Biomedical Engineering Program, University of Calgary, Calgary, AB, Canada, ²Departments of Radiology and Clinical Neurosciences, University of Calgary, Calgary, AB, Canada

Standard MRI is routinely collected in patient care but is limited in assessing changes in tissue microstructure. We developed a new method to assess tissue directionality using the power spectrum of T2-weighted MRI and validated it using the highly coherent structure, corpus callosum. In controls, power spectrum-derived angles corresponded exactly with the predicted aligning directions of the corpus callosum, and such aligning patterns were interrupted in advanced MS patients with increased variability and angular entropy. Fourier-based power spectrum may provide advanced measures of tissue directionality following myelin and axonal pathology using clinical scans.

Doaa Mousa¹, Nourhan Zayed¹, and Inas Yassine^2,3
1Computer and Systems, Electronic Research Institute, Giza, Egypt, ²Systems and Biomedical Engineering, Cairo University, Giza, Egypt, ³Medical Informatics and Image processing Lab, Nile University, Giza, Egypt

Cardiovascular diseases (CVDs) cause 31% of the death rate globally. Automatic accurate segmentation is needed for CVDs early detection. In this paper, we propose a modified workflow to automatically segment the left ventricle (LV) for the short axis cardiac perfusion MRI (perfusion CMR) images using levelset method. We propose mitigating the initial contour extraction, and modify the technique used to initialize the levelset algorithm in order to improve the accuracy of segmentation results. The system workflow consists of five main modules: preprocessing, localization, initial contour extraction, registration, and segmentation. Our results showed enhancement in the segmentation accuracy by 5%.

Han Sang Lee¹ and Junmo Kim^1
1School of Electrical Engineering, KAIST, Daejeon, Korea, Republic of

Multi-atlas segmentation has often suffered from the registration error. We propose a novel method for multi-atlas registration for multi-atlas segmentation inspired by the template generation and deep neural network. We first add an intra-atlas registration layer which performs image-based registration between atlas images to duplicate the atlases. We then add a label-wise registration layer which rectifies the registered images by label-based registration. We present preliminary results of our multi-layered atlas registration on brain MRI segmentation.

Michael Amann^1,2, Pavel Falkovskiy^3,4,5, Alain Thoeni¹, Tobias Kober^3,4,5, Alexis Roche^3,4,5, Bénédicte Maréchal^3,4,5, Philippe Cattin⁶, Tobias Heye², Oliver Bieri², Till Sprenger⁷, Christoph Stippich², Gunnar Krueger^4,5,8, Ernst-Wilhelm Radue¹, and Jens Wuerfel^1
1Medical Image Analysis Center (MIAC), Basel, Switzerland, ²Department of Radiology, University Hospital of Basel, Basel, Switzerland, ³Advanced Clinical Imaging Technology, Siemens Healthcare AG, Lausanne, Switzerland, ⁴Department of Radiology, University Hospital (CHUV), Lausanne, Switzerland, ⁵École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, ⁶Department of Biomedical Engineering, University of Basel, Basel, Switzerland, ⁷Department of Neurology, DKD Helios Klinik, Wiesbaden, Germany, ⁸Siemens Medical Solutions USA, Boston, MA, United States

Performance of FreeSurfer and FSL was compared on T1-weighted 3D MRI data of 22 controls as function of scan session, scanner type and segmentation pipeline. Intra-class correlation coefficients and percentage volume differences were calculated for the segmentation results of both pipelines. Strong agreement was found for whole brain, white matter and cortex. For each pipeline, the impact of experimental factors was assessed by linear mixed effects analysis. We found significant scanner effect on the results of both segmentation pipelines. For subcortical structures, segmentation reliability was higher in FSL than in FreeSurfer, whereas for cortex and WM, FreeSurfer was more stable.

Renjie He¹, Yu Ding¹, and Qi Liu^1
1United Imaging Healthcare America, Houston, TX, United States

Geometric average is insensitive to the value variation between components to be averaged, this is used to noticeably reduce the inhomogeneity caused by Sum-of-Squares (SOS) in channel combination in parallel MR imaging.

Antoine Lutti^1
1LREN, Dept. of Clinical Neurosciences, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland

The characterization of brain microstructure from MRI data requires the development of specific MRI tissue biomarkers and of advanced models linking microscopic tissue properties to MRI signals. We apply the Anderson-Weiss theory, which describes the transverse relaxation of the MRI signal as a function of tissue microstructure, on in-vivo MRI data acquired at 3T. In grey matter, parameter estimates show a strong correlation with histological measures of iron concentration. The time constants provided by the model yield realistic estimates of microscopic compartment size. These results offer a promising perspective for the histological assessment of brain tissue in-vivo using MRI.

Moira Marizzoni¹, Daniele Orlandi¹, Luigi Antelmi², Flavio Nobili³, Mira Didic^4,5, David Bartrés-Faz⁶, Ute Fiedler⁷, Peter Schonknecht⁸, Pierre Payoux^9,10, Andrea Soricelli^11,12, Alberto Beltramello¹³, Lucilla Parnetti¹⁴, Magda Tsolaki¹⁵, Paolo Maria Rossini^16,17, Pieter Jelle Visser¹⁸, Regis Bordet¹⁹, Oliver Blin²⁰, Giovanni Battista Frisoni^1,21, Jorge Jovicich²², and on behalf of the PharmaCog Consortium^1
1LENITEM Laboratory of Epidemiology, Neuroimaging, & Telemedicine — IRCCS San Giovanni di Dio-FBF, Brescia, Italy, ²Health Department, Foundation IRCCS Neurological Institute Carlo Besta, Milan, Italy,³Department of Neuroscience, Ophthalmology, Genetics and Mother–Child Health (DINOGMI), University of Genoa, Genoa, Italy, ⁴APHM, CHU Timone, Service de Neurologie et Neuropsychologie, Marseille, France,⁵Aix-Marseille Université, INSERM U 1106, Marseille, France, ⁶Department of Psychiatry and Clinical Psychobiology, Universitat de Barcelona and IDIBAPS, Barcelona, Spain, ⁷LVR-Clinic for Psychiatry and Psychotherapy, Institutes and Clinics of the University Duisburg-Essen, Essen, Germany, ⁸Department of Neuroradiology, University Hospital Leipzig, Leipzig, Germany, ⁹INSERM, Imagerie cérébrale et handicaps neurologiques, UMR 825, Toulouse, France, ¹⁰Université de Toulouse, UPS, Imagerie cérébrale et handicaps neurologiques, UMR 825, CHU Purpan, Place du Dr Baylac, Toulouse, France, ¹¹IRCCS SDN, Naples, Italy,¹²University of Naples Parthenope, Naples, Italy, ¹³Department of Neuroradiology, General Hospital, Verona, Italy, ¹⁴Section of Neurology, Centre for Memory Disturbances, University of Perugia, Perugia, Italy, ¹⁵3rd Department of Neurology, Aristotle University of Thessaloniki, Thessaloniki, Greece, ¹⁶Dept. Geriatrics, Neuroscience & Orthopaedics, Catholic University, Policlinic Gemelli, Rome, Italy, ¹⁷IRCSS S.Raffaele Pisana, Rome, Italy, ¹⁸Department of Neurology, Alzheimer Centre, VU Medical Centre, Amsterdam, Italy, ¹⁹Department of Pharmacology, EA1046, University of Lille Nord de France, Lille, Italy, ²⁰Pharmacology, Assistance Publique-Hôpitaux de Marseille, Aix-Marseille University-CNRS UMR 7289, Marseille, France, ²¹Memory Clinic and LANVIE - Laboratory of Neuroimaging of Aging, University Hospitals and University of Geneva, Geneva, Switzerland, ²²Center for Mind/Brain Sciences, University of Trento, Rovereto, Italy

In this study we quantify the across-session reproducibility of hippocampus subfields obtained from the recently proposed ex-vivo atlas tool available in Freesurfer version 6.0. We use structural 3T multisite data from 65 healthy elderly participants scanned twice at least a week apart. We show that several subfields  like Cornu Ammonis (CA) 1, hippocampal tail, molecular layer and subiculum offer, despite being smaller, comparable reliability errors to the whole hippocampus volume (2%). This suggests that these subfields may be valid and more specific markers to test disease progression in longitudinal studies, like for example Alzheimer's disease.

Shunshan Li¹, Lily Zhou², Mark J Fisher³, Ronald C Kim⁴, Vitaly Vasilevko⁵, David Cribbs⁵, Annlia Hill³, and Min-Ying Su^6
1Tu & Yuen Center for Functional Onco-Imaging, Department of Radiological Sciences, university of california, irvine, irvine, CA, United States, ²Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University, Guangzhou, China, People's Republic of, ³Department of Neurology, University of California, Irvine, Irvine, CA, United States, ⁴Department of Pathology, University of California, Irvine, Irvine, CA, United States, ⁵Institute for Memory Impairments and Neurological Disorders, University of California, Irvine, irvine, CA, United States, ⁶Tu & Yuen Center for Functional Onco-Imaging, Department of Radiological Sciences, University of California, Irvine, irvine, CA, United States

The postmortem brain MR images include air-bubble artifacts and typical microbleeds(MBs) are less than 200 µm which make MBs detection very challenging. In this project we developed an optimization MR imaging method to detect possible MBs on postmortem brains of patients with and without dementia, hoping to provide information to guide neuropathological examination to sample the suspicious MBs areas, and improve the chance of identifying true MBs to better understand its role in normal aging and development/progression of dementia, and further develop streamlined automatic MBs detection software.

Sameeha Fallatah¹, Rolf Jäger¹, and Xavier Golay^1
1Brain Repair and Rehabilitation, UCL, Institute of Neurology, London, United Kingdom

Dynamic Contrast Enhanced MRI is used to assess the integrity of the blood brain barrier. A major difficulty for the method to be accepted in the clinics is the variety of pharmacokinetic models used and their strong dependence on the underlying assumptions and/or acquisition parameters. Thus the far simpler methods based on signal intensity curve characteristics are the most commonly used approaches in clinical practice. In this study we compare two different pharmacokinetic models, the extended Tofts model and Lawrence & Lee model in patients with primary brain tumours. 

Gaurav Verma¹, Suyash Mohan¹, Sanjeev Chawla¹, John Y.K. Lee², Sumei Wang¹, Andrew Maudsley³, Steven Brem², and Harish Poptani^4
1Department of Neuroradiology, University of Pennsylvania, Philadelphia, PA, United States, ²Department of Neurosurgery, University of Pennsylvania, Philadelphia, PA, United States, ³Department of Radiology, University of Miami, Miami, FL, United States, ⁴Department of Cellular and Molecular Physiology, University of Liverpool, Liverpool, United Kingdom

A 3D visualization, co-registration and quantification platform was developed in Matlab to combine anatomical imaging with physiological and metabolic data from diffusion tensor, perfusion-weighted and echo-planar spectroscopic imaging. This data can be co-registered across modalities and imaging time-points to provide detailed information about the spatial extent of a brain tumor. 3D visualization was applied in datasets from patients undergoing neurosurgery and a separate cohort of patients undergoing long-term Tumor Treating Fields (TTFields) therapy. This visualization platform could have an impact in the planning of neurosurgery and the placement and monitoring of location-sensitive techniques like TTFields.

David Bondesson^1,2, Thomas Gaass^1,3, Julien Dinkel^1,2, and Berthold Kiefer^4
1Josef Lissner Laboratory for Biomedical Imaging, Department of Clinical Radiology, Ludwig-Maximilians-University Hospital Munich, Munich, Germany, ²Comprehensive Pneumology Center, German Center for Lung Research, Munich, Germany, ³Comprehensive Pneumology Center,German Center for Lung Research, Munich, Germany, ⁴Siemens AG Healthcare Sector, Erlangen, Germany

Evaluating regional lung perfusion and ventilation is diagnostically valuable in regards of pulmonary diseases. Standard methods however, expose patients to risks from ionizing radiation and contrast agents. MRI screening is not based on radiation and a new method has previously been presented as a non-contrast-enhanced estimation. This work presents wavelet decomposition as a potential improvement to fourier decomposition for perfusion and ventilation assessment of the human lung in proton MRI.  

Sudhakar Tummala¹, Bumhee Park¹, Ruchi Vig¹, Mary A Woo², Daniel W Kang³, Ronald M Harper^4,5, and Rajesh Kumar^1,5,6,7
1Anesthesiology, University of California at Los Angeles, Los Angeles, CA, United States, ²UCLA School of Nursing, Los Angeles, CA, United States, ³Medicine, University of California at Los Angeles, Los Angeles, CA, United States, ⁴Neurobiology, University of California at Los Angeles, Los Angeles, CA, United States, ⁵Brain Research Institute, University of California at Los Angeles, Los Angeles, CA, United States, ⁶Radiological Sciences, University of California at Los Angeles, Los Angeles, CA, United States, ⁷Bioengineering, University of California at Los Angeles, Los Angeles, CA, United States

Obstructive sleep apnea subjects show gray matter volume loss in multiple brain areas, based on voxel-based morphometry procedures, which are less sensitive in detecting subtle chronic/acute gray or white matter changes. We assessed brain injury in recently-diagnosed, treatment naïve OSA subjects by evaluating regional entropy, which measures the extent of homogeneity or randomness in tissue texture, and found significantly decreased regional entropy values in areas regulating autonomic, respiratory, cognitive, and neuropsychologic functions that are deficient in the condition, suggesting predominantly acute tissue pathology in those sites.. The findings suggest that regional entropy can demonstrate acute tissue changes.

Frank Zijlstra¹, Job G Bouwman¹, Ieva Braškute¹, and Peter R Seevinck^1
1Image Sciences Institute, UMC Utrecht, Utrecht, Netherlands

We present a fast alternative to Bloch simulation for simulation of off-resonance artifacts in steady-state imaging. By assuming a steady-state, the signal equation can be quickly evaluated by using multiple Fast Fourier Transforms. We show an acceleration factor of over 350 for a 2D simulation of a titanium cylinder phantom, while the differences with Bloch simulation were minor. The speed of the proposed method enables 3D simulations at high resolution and may benefit various applications.

Minju Jo¹, Yoonho Nam², Jeehun Kim¹, Hyeong Geol Shin¹, and Jongho Lee^1
1Laboratory for Imaging Science and Technology, Department of Electrical and Computer Engineering, Seoul National University, Seoul, Korea, Republic of, ²Department of Radiology, Seoul St. Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea, Republic of

In this work, we present a method of estimating the phase offsets in multi-echo GRE data, Multi-Channel Phase Combination using all N echoes (MCPC-N). MCPC-N, which calculates the phase offsets from all echoes, provides more accurate estimation of voxel-wise phase offsets particularly in low SNR.

Eric James Keller¹, Jeremy Douglas Collins¹, Cynthia K Rigsby², James C Carr¹, Michael Markl^1,3, and Susanne Schnell^1
1Radiology, Northwestern University, Chicago, IL, United States, ²Radiology, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, IL, United States, ³Biomedical Engineering, Northwestern University, Evanston, IL, United States

4D flow MRI quantification of abdominal hemodynamics is challenged by a wide range of blood flow velocities and vessel diameters. By adjusting critical pre-processing steps required to analyze 4D flow MRI data, we were able to both recover vessels of interest lost by our previous method and significantly reduce the relative error in flow measurements. We conclude that it is critical to apply background phase error correction prior to any other filters and/or corrections to ensure accurate background offset estimation. Additionally, low venc acquisitions should not be noise corrected to ensure low flow data is not inadvertently deleted.

Lixin Yang¹, Wei Yu¹, Zhaoyang Fan², and DeBiao Li^3
1Department of Radiology, Beijing AnZhen Hospital, Beijing, China, People's Republic of, ²Biomedical Imaging Research Institute, Cedars-Sinai Medical Center, Los Angeles, CA, United States, ³Biomedical Imaging Research Institute, Cedars-Sinai Medical Center,Department of Bioengineering, University of California, Los Angeles, CA, United States

Purpose: Determine the accuracy of MATCH in the characterization of plaque composition in patients in comparison with the conventional multi-contrast approach, using histopathology as the gold standard.

Methods: Twenty-two patients scheduled for carotid endarterectomy underwent preoperative carotid MRI with MATCH and the conventional protocol, blinded image review for composition identification was performed by 2 radiologists. Carotid histopathological specimens stained with HE and Masson, matched with this two protocol images, Cohen kappa (K) was computed to quantify the agreement in the detection of components among this two protocols and histopathology.

Results: Moderate to good agreement was seen between histopathological specimens and multi-contrast protocol in the detection of plaque components (IH k=0.704 , CA k=0.763, LR/NC k=0.844). Similar results were seen between histopathological specimens and MATCH (IH k=0.703CA k=0.740, LR/NC k=0.850). 

Kemal Sümser¹, Nashwan Naji^1,2, Mehdi Sadighi¹, Hasan Hüseyin Eroglu^1,3, and Murat Eyüboglu^1
1Electrical and Electronics Engineering Department, Middle East Technical University, Ankara, Turkey, ²On Leave from Ibb University, Ibb, Yemen, ³TSK Rehabilitation and Care Center, Ankara, Turkey

In magnetic resonance electrical impedance tomography (MREIT) currents are injected to the object during MRI imaging sequence. In this study, we propose a new pulse sequence based on the spatial modulation of magnetization (SPAMM) to be used in MREIT applications. In this pulse sequence, the current is injected during a pre SPAMM module which can be followed by any conventional Magnetic Resonance Imaging pulse sequence for data acquisition. Experimental result in comparison with the simulation result shows that this method is an applicable technique for MREIT data acquisition.

Endre Grøvik^1,2, Atle Bjørnerud^1,2, Tryggve Holck Storås¹, Kjell-Inge Gjesdal³, and Kathinka Dæhli Kurz^4
1The Intervention Centre, Oslo University Hospital, Oslo, Norway, ²Department of Physics, University of Oslo, Oslo, Norway, ³Sunnmøre MR klinikk AS, Ålesund, Norway, ⁴Department of Radiology, Stavanger University Hospital, Stavanger, Norway

The purpose was to evaluate the influence of BAT in pharmacokinetic analysis of breast masses, by estimating the kinetic parameters both with and without BAT-delay correction. Thirty-nine verified breast masses were examined using a high temporal resolution EPI sequence. The image-data were analyzed using a two-compartment kinetic model with and without BAT-delay correction. The relationship between the relative parametric error and BAT-delay were investigated. The result indicates that neglecting the delayed BAT leads to an overestimation of K^trans, k_ep, and, v_e, and a underestimation of v_p, and that the delayed BAT needs to be accounted for in the model-based analysis.

Emily Etchell¹, Lauriane Jugé ^1,2, Alice Hatt¹, Ralph Sinkus³, and Lynne E. Bilston^1,4
1Neuroscience Research Australia, Randwick, NSW, Australia, ²School of Medical Sciences, University of New South Wales, Kensington, NSW, Australia, ³BHF Centre of Excellence, Division of Imaging Sciences and Biomedical Engineering, King's College London, London, United Kingdom, ⁴Prince of Wales Clinical School, University of New South Wales, Kensington, NSW, Australia

Magnetic resonance (MR) elastography provides clinical information for chronic hepatic disorders by quantifying an increase in liver stiffness compared to heathy baseline values. Thus far however, baseline stiffness values have only been reported for adults. We aimed to fill this gap by quantifying healthy liver stiffness of children and adolescents. Results showed that pediatric liver stiffness increases with age during normal development, approaching adult values during adolescence. This implies that comparing pediatric liver stiffness measurements to adult baseline values when using MR elastography may miss disease or underestimate disease severity.

Prateek Kalra¹, Arunark Kolipaka, PhD¹, Jeffrey R. Hawley, MD¹, and Brian Raterman^1
1Radiology, Ohio State University Wexner Medical Center, Columbus, OH, United States

Magnetic resonance elastography (MRE) is a non-invasive technique to estimate stiffness of soft tissues and has been applied in the breast. However, none of the earlier studies have extensively tested MRE to induce vibrations in the breast using a soft sternum driver at higher field strength and its repeatability of stiffness measurements. The aim of the study is to estimate breast stiffness using MRE by inducing vibrations using a soft sternum driver in normal volunteers at 3T and to determine the reproducibility of stiffness measurements. Preliminary results show that the MRE-derived stiffness values are reproducible in normal volunteers at 3T and can be further extended to detect breast tumors in patients. 

Andreas Fehlner¹, Sebastian Hirsch¹, Mykola Kadobianskyi², Patric Birr¹, Eric Barnhill^1,3, Martin Weygandt^2,4, Johannes Bernarding⁵, Jürgen Braun⁶, Ingolf Sack¹, and Stefan Hetzer^2,4
1Department of Radiology, Charité - Universitätsmedizin Berlin, Berlin, Germany, ²Berlin Center for Advanced Neuroimaging, Charité - Universitätsmedizin Berlin, Berlin, Germany, ³Clinical Research Imaging Centre, School of Clinical Sciences and Community Health, College of Medicine and Veterinary Medicine, The University of Edinburgh, Edinburgh, United Kingdom, ⁴Bernstein Center for Computational Neuroscience Berlin, Berlin, Germany, ⁵Institut für Biometrie und Medizinische Informatik, Universitätsklinikum Magdeburg, Magdeburg, Germany, ⁶Institute of Medical Informatics, Charité - Universitätsmedizin Berlin, Berlin, Germany

High-resolution Multifrequency MR Elastography (MMRE) is hampered by susceptibility-induced image distortions. We corrected MMRE data of a 3T and 7T MR scanner for motion and EPI distortion artefacts. The correction of subject motion significantly sharpened the images, which was demonstrated by a decrease of the point-spread function. The improvement was highly correlated with the degree of subject motion. Distortion correction enhanced the accuracy of normalization in the MNI152 space as shown by an increase of the correlation between individual and standard tissue probability maps. This method could help increasing the sensitivity of multi-subject studies exploring |G*| e.g. in small subcortical areas.

Florian Dittmann¹, Heiko Tzschätzsch¹, Jing Guo¹, Sebastian Hirsch¹, Jürgen Braun², and Ingolf Sack^1
1Institute of Radiology, Charité, Berlin, Germany, ²Department of Medical Informatics, Charité, Berlin, Germany

We demonstrate the feasibility of in vivo prostate exam utilizing shear waves induced by pressurized-air actuators previously developed for abdominal MRE.  High wave amplitudes throughout the prostate were achieved in the lower frequency regime from 30 to 50 Hz. Using a 2D multifrequency wave number inversion algorithm, wave speed maps with sufficiently high resolution are obtained to discriminate between the central zone and peripheral zone despite longer wavelengths pertaining to lower vibration frequencies. The proposed MRE setup promises robust and easy-to-use applications in the clinic without the need of specialized hardware in addition to the abdominal MRE setup.

Joshua D. Trzasko¹, Arvin Arani¹, Armando Manduca¹, Kevin J. Glaser¹, Richard L. Ehman¹, Philip A . Araoz¹, and John Huston III^1
1Mayo Clinic, Rochester, MN, United States

In this work, we adapt the previously-described robust harmonic estimation (RHE) strategy for magnetic resonance elastography (MRE) to brain imaging, and demonstrate that use of this novel signal processing tool improves the accuracy of estimated stiffness information both in a geometrically-accurate phantom and in vivo.

Curtis L Johnson¹, Emily S Matijevich^1,2, Emily D Cullum^1,2, Matthew DJ McGarry³, Keith D Paulsen³, Bradley P Sutton^1,4, Tracey M Wszalek^1,2, and William C Olivero^1,2
1Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ²Carle Neuroscience Institute, Carle Foundation Hospital, Urbana, IL, United States,³Thayer School of Engineering, Dartmouth College, Hanover, NH, United States, ⁴Department of Bioengineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States

MR elastography has emerged as an important tool for presurgical evaluation of intracranial tumors. Due to the localized nature of the lesion properties of interest, there is a need for high-resolution MRE methods for characterizing tumors. Here we present our initial experience with MRE of intracranial tumors using a protocol based on high-resolution imaging and nonlinear inversion. We found that glial tumors are soft and have a generally low viscosity, while meningeal tumors are stiff and have a very low viscosity.

Florian Dittmann¹, Sebastian Hirsch¹, Jing Guo¹, Jürgen Braun², and Ingolf Sack^1
1Institute of Radiology, Charité, Berlin, Germany, ²Department of Medical Informatics, Charité, Berlin, Germany

Since shear waves at low drive frequencies are nearly unaffected by attenuation, we introduce a brain MRE setup, which is based on remote excitation of intracranial shear waves by a pressurized-air actuator in the regime of 20 Hz. MRE-scans, which were repeated 27 times on three different days for each of six healthy volunteers, show differences between individuals as well as from day-to-day for the same individual. The investigation demonstrates that cerebral low frequency MRE provides a fast and reproducible novel source of mechanical information of brain tissue with less onerous head stimulation as required by conventional MRE.

Suraj D Serai¹, Jonathan R Dillman¹, Hui Wang², and Andrew T Trout^1
1Radiology, Cincinnati Children's Hospital, Cincinnati, OH, United States, ²Philips Healthcare, Cincinnati, OH, United States

MR elastography (MRE) allows non-invasive evaluation of hepatic stiffness and samples a larger area of the liver than liver biopsy. The high accuracy of MRE for liver fibrosis staging suggests that MRE could potentially replace liver biopsy. MRE has traditionally been performed using a GRE sequence. GRE, however, has SNR limitations at higher field strengths that can result in under-sampling potentially leading to erroneous stiffness values. SE-EPI is an alternative means of performing MRE that has higher SNR, lower susceptibility related signal loss and increased speed. In this work, we compared GRE and SE-EPI MRE across two vendor platforms.  

Jin Li¹, Lisa Asher¹, Filipa Lopes², Craig Cummings¹, Alexander Koers^2,3, Laura S. Danielson^2,3, Louis Chesler^2,3, Caroline J. Springer², Jeffrey C. Bamber¹, Ralph Sinkus⁴, Yann Jamin¹, and Simon P. Robinson^1
1Division of Radiotherapy & Imaging, The Institute of Cancer Research, London, United Kingdom, ²Division of Cancer Therapeutics, The Institute of Cancer Research, London, United Kingdom, ³Division of Clinical Studies, The Institute of Cancer Research, London, United Kingdom, ⁴Division of Imaging Sciences and Biomedical Engineering, King’s College London, King’s Health Partners, St. Thomas’ Hospital, London, United Kingdom

MRE was applied to assess the viscoelastic properties of orthotopic pancreatic ductal adenocarcinoma (PDAC) xenografts, and tumours arising in a transgenic mouse model of MYCN-amplified neuroblastoma, within the mouse abdomen. The stromal-rich PDAC tumours were quantified with markedly elevated elasticity (G_d) and viscosity (G_l), whilst the pathologically diverse neuroblastomas exhibit more heterogeneity in their biomechanical properties and were relatively soft. MRE can non-invasively assess the viscoelastic properties of deep-seated tumours arising within the abdomen of mice in vivo. 

Ian Gavin Murphy¹, Martin Graves², Scott Reid³, Andrew Patterson², Ilse Gavin Joubert¹, Andrew N Priest², and David J Lomas^2
1Radiology, Cambridge University Hospitals NHS Foundation Trust, Cambridge, United Kingdom, ²Radiology, Cambridge University NHS Foundation Trust, Cambridge, United Kingdom, ³GE, Little Chafont, United Kingdom

In patients with liver disease, MR elastography (MRE) is a non-invasive method for evaluating fibrosis. MRE is phase-based and sensitive to motion artefact, and is typically performed in end expiration. We found that navigator timed MRE shows no statistical difference to breath-held techniques for stiffness and reproducibility in 6 healthy volunteers, and may prove superior in patients unable to adequately hold their breath 

Curtis L Johnson¹, Joseph L Holtrop^1,2, Aaron T Anderson³, and Bradley P Sutton^1,2
1Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ²Department of Bioengineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ³Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States

We propose a novel sequence for magnetic resonance elastography (MRE) of the brain based on multiband excitation and 3D encoding of the distributed slab with multishot spirals. This sequence allows access to optimal SNR efficiency and reduced distortions from field inhomogeneity, but also parallel imaging acceleration both in-plane and thru-plane without onerous artifacts and g-factor penalties. We also incorporate correction for nonlinear motion-induced phase errors through a k_z-blipped spiral-in 3D navigator. In this abstract we demonstrate the performance of the sequence and its ability to capture whole-brain MRE data at 2x2x2 mm³ resolution in 3 minutes.

Eric Barnhill¹, Jing Guo², Florian Dittmann², Sebastian Hirsch², Michael Perrins³, Lucy Hiscox³, Tim Herrmann⁴, Johannes Bernarding⁴, Neil Roberts³, Jürgen Braun¹, and Ingolf Sack^2
1Institute of Medical Informatics, Charité Universitätsmedizin Berlin, Berlin, Germany, ²Department of Radiology, Charité Universitätsmedizin Berlin, Berlin, Germany, ³Clinical Research Imaging Centre, The University of Edinburgh, Edinburgh, United Kingdom, ⁴Institute for Biometrics and Medical Informatics, Otto von Guericke University Magdeburg, Magdeburg, Germany

We investigated the impact of field strength and image resolution on brain MRE stiffness results.  A cohort of 18 healthy volunteer subjects was scanned at 1.5T (2mm isotropic voxels), 3T (2mm) and 7T (1mm), with a fourth set downsampling the 7T to 2mm.  Means were 1634 Pa (+/-613) for 1.5T, 1743 Pa (+/-811) for 3T, 1786 (+/-634) Pa for 7T 2mm, and 927 (+/-364) Pa for 7T 1mm. In the paired sign-rank tests, there were no significant effects for field strength. Examination of histograms of example slices suggests that a different distribution of features is being captured at the higher resolution.

Eric Barnhill¹, Florian Dittmann², Sebastian Hirsch², Jing Guo², Jürgen Braun¹, and Ingolf Sack^2
1Institute of Medical Informatics, Charité Universitätsmedizin Berlin, Berlin, Germany, ²Department of Radiology, Charité Universitätsmedizin Berlin, Berlin, Germany

Magnetic Resonance Elastography (MRE) stiffness estimates show differentiated results by feature scale. Here progressive denoising was applied to study the relation between image sharpness (as meaured by Reduced Energy Ratio) and image stiffness estimate (as measured by complex shear modulus magnitude |G*|). Progressive complex-wavelet-based denoising appears to reach stable stiffness estimates in phantom and brain acquisitions. Images of maximum sharpness result in lower overall stiffness estimates than the stable global estimate, suggesting that coarse elasticity estimates do not average fine feature results, but measure a different stiffness scale.

Roger Grimm¹, Jun Chen¹, and Richard Ehman^1
1Mayo Clinic, Rochester, MN, United States

A 3D gradient recalled echo sequence has been developed that samples the three shear wave displacement polarization at 3 time points for a total of 9 image samples. A multi-coil recon generates phase difference images and then uses a 3 point discrete Fourier transform to provide the complex displacement fields. The sequence is shown in breast and head applications.

Yohei Itoh¹, Yasuo Takehara², Naoki Ooishi², Masanori Kawade², Tetsuya Wakayama³, Mikio Suga⁴, Takasuke Ushio¹, Yuki Hirai¹, Nobuko Yoshizawa¹, Shuhei Yamashita¹, Hatsuko Nasu¹, and Harumi Sakahara^1
1Diagnostic Radiology & Nuclear Medicine, Hamamatsu University school of medicine, Hamamatsu, Shizuoka, Japan, ²Department of Radiology, Hamamatsu University Hospital, Hamamatsu, Shizuoka, Japan,³GEHCJ, Hino-shi, Tokyo, Japan, ⁴Center for Frontier Medical Engineering, Chiba University, Chiba, Japan

To achieve the high spatial resolution MR elastography(MRE), it has been reported that combining SS-SE-EPI with a spatially selective excitation provides an efficient way of reducing FOV. Using this technique, we performed phantom study and pancreatic MRE. We report the result of our study and the point we found when performing the reduced-FOV MRE.

Heiko Tzschätzsch¹, Jing Guo¹, Florian Dittmann¹, Sebastian Hirsch¹, Eric Barnhill¹, Jürgen Braun², and Ingolf Sack^1
1Department of Radiology, Charité - University Medicine Berlin, Berlin, Germany, ²Institute of Medical Informatics, Charité - University Medicine Berlin, Berlin, Germany

Elastography often suffers from limited anatomical resolution due to noise and insufficient elastic deformation. We here introduce noise-robust multifrequency wave number inversion for multifrequency MR elastography. Compound maps of wave speed are obtained, which reveal variations in tissue elasticity in a tomographic fashion, i.e. an unmasked, slice-wise display of anatomical details at pixel-wise resolution. The method is demonstrated using data from the literature including abdominal and pelvic organs such as the liver, spleen, uterus and cervix. Elastic parameters consistent with literature values were obtained even in small regions with low wave amplitudes such as nucleus pulposus and spinal cord.

Navid Samavati^1,2, Clara Körting¹, Toni Drießle³, Stefan Wintzheimer³, Jing Guo¹, Florian Dittmann¹, Ingolf Sack¹, and Jürgen Braun^2
1Department of Radiology, Charité University Medicine, Berlin, Germany, ²Department of Medical Informatics, Charité University Medicine, Berlin, Germany, ³Pure Devices GmbH, Würzburg, Germany

A fully integrated tabletop MR elastography (MRE) system based on a 0.5-T permanent magnet for investigations of small tissue samples is introduced. A 3D spin echo MRE sequence allows control of all MRE parameters including frequency and amplitude of a piezoelectric actuator. The device enables fully automated measurements of maps of viscoelastic parameters in soft tissue samples by 3D multifrequency MRE. Initial results are in good agreement to published data and demonstrate the great potential of the system as a preclinical research unit in histopathological laboratories and operating rooms.

Lumeng Cui^1,2, Conrad Yuen³, Ted Lynch⁴, Paul Babyn⁵, Francis M. Bui⁶, and Niranjan Venugopal^1,5
1Department of Medical Physics, Saskatchewan Cancer Agency, Saskatoon, SK, Canada, ²Division of Biomedical Engineering, University of Saskatchewan, Saskatoon, SK, Canada, ³Department of Medical Physics, BC Cancer Agency, Vancouver, BC, Canada, ⁴Non-ionizing Radiation, CIRS Inc., Norfolk, VA, United States, ⁵Department of Medical Imaging, University of Saskatchewan, Saskatoon, SK, Canada, ⁶Department of Electrical & Computer Engineering, University of Saskatchewan, Saskatoon, SK, Canada

Magnetic Resonance Elastography (MRE) is a new imaging technique that combines the acoustic waves and MRI to retrieve elastic properties of tissue. Because MRE is non-invasive, there is great clinical interest for its use in the detection of cancer. In this work, we focus on the design of an MRE phantom to be used in the clinical commissioning of an MRE System. With the aid of newly designed pulse sequences and inversion algorithms we have developed a quality assurance process to validate the efficacy of MRE for applications to many clinical sites (i.e. prostate, cervix, uterus).

Renee Miller¹, Arunark Kolipaka², Vicky Wang³, Martyn Nash³, and Alistair Young^1
1Anatomy with Radiology, University of Auckland, Auckland, New Zealand, ²Radiology, Biomedical Engineering and Internal Medicine, The Ohio State University Wexner Medical Center, Columbus, OH, United States,³Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand

In this study, we examined the determinability of anisotropic stiffness parameters using finite element analysis simulations of harmonic steady-state wave behaviour. Two simulation experiments, of cylindrical phantom and left ventricular geometries, and one phantom experiment using magnetic resonance elastography (MRE) were carried out. The transversely isotropic material properties were determined using a least-squares optimisation algorithm by matching a modelled displacement field to the reference, or MRE, displacement field. The results showed that the parameters were uniquely identifiable even in the presence of noise.

Cemre Ariyurek^1,2, Safa Ozdemir^1,2, Arif Sanli Ergun³, Yusuf Ziya Ider¹, and Ergin Atalar^1,2
1Department of Electrical and Electronics Engineering, Bilkent University, Ankara, Turkey, ²National Magnetic Resonance Research Center (UMRAM), Ankara, Turkey, ³Department of Electrical and Electronics Engineering, TOBB-University of Economics and Technology, Ankara, Turkey

Experimental validation of modes of shear waves in MR elastography (MRE) is demonstrated. For the first time, frequency response of the actuator is investigated and actuator displacement is measured. Normalizing shear wave displacement to the actuator displacement removes the effect of actuation system and isolates shear wave resonance. It is demonstrated that 10-20 times greater shear wave displacement than applied displacement by the actuator can be observed at resonance. Thus, safety issues in MRE should be reconsidered. Presenting repeatability of determining mode frequency validates feasibility of detecting stiffness changes by observing any shift in mode frequency.

Ligin Solamen¹, Matthew McGarry¹, Elijah Van Houten², Jennifer Hong³, John Weaver^1,4, and Keith Paulsen^1,5
1Thayer School of Engineering, Dartmouth College, Hanover, NH, United States, ²Department of Mechanical Engineering, University of Sherbrooke, Sherbrooke, QC, Canada, ³Department of Neurosurgery, Dartmouth-Hitchcock Medical Center, Lebanon, NH, United States, ⁴Department of Radioogy, Dartmouth-Hitchcock Medical Center, Lebanon, NH, United States, ⁵Norris Cotton Cancer Center, Dartmouth-Hitchcock Medical Center, Lebanon, NH, United States

Intrinsically actuated poroelastic MR elastography (IA-pMRE) is a technique which estimates tissue mechanical and hydrodynamic properties using measurements of displacement during the cardiac cycle, and does not require external vibration as in traditional MRE. Compared to conventional MRE, which obtains displacements in the range of 25-100Hz, IA-pMRE uses intrinsically generated low frequency (1-2Hz) displacements for elastography reconstruction. IA-pMRE was applied to 7 brain tumor patients and showed a significant difference in both the shear modulus and hydraulic conductivity of brain tissue compared to healthy tissue.