Mo Shahdloo¹, Daniel Papp², Urs Schüffelgen¹, Karla L. Miller², Matthew Rushworth¹, and Mark Chiew^2
1Wellcome Centre for Integrative Neuroimaging, Department of Experimental Psychology, University of Oxford, Oxford, United Kingdom, ²Wellcome Centre for Integrative Neuroimaging, FMRIB, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, United Kingdom

Dynamic B0 field inhomogeneities due to vigorous body motion - after mechanical head stabilisation - introduce severe artefacts in accelerated fMRI of awake behaving non-human primates (NHPs) by invalidating the calibration data used for unaliasing reconstruction. Here, we propose a method to estimate dynamic field perturbations via model-based frame-to-frame comparison of EPI reference navigators. These estimates can be used to improve reconstruction quality by matching each data frame to the calibration data, and simultaneously correcting the geometric distortions. The proposed method successfully estimates field perturbations and improves reconstruction quality in accelerated NHP fMRI, without the need for sequence modification or extra acquisitions.

Faisal Najeeb¹, Jichang Zhang², Xinpei Wang², Chengbo Wang², Hammad Omer¹, Penny Gowland³, Sue Francis³, and Paul Glover^3
1MIPRG,Comsats University, Islamabad, Pakistan, ²SPMIC, The University of Nottingham Ningbo China, Ningbo, China, ³SPMIC, The University of Nottingham, Nottingham, United Kingdom

In this work,  temporal shift-windowing function is integrated into Low rank plus sparse (L+S) decomposition to fix the certain motion phase with a high temporal resolution and reconstruction efficiency for free-breathing golden angle radial DCE-MRI of liver. A smooth weighting curve based on a sigmoid function is used to achieve a smooth transition for the spokes between the desired phase and other motion phases. Furthermore, the Fast Iterative Shrinkage-thresholding Algorithm (FISTA) was implemented to solve the L+S optimization problem which enables faster convergence. Results of the proposed method are  compared with RACER-GRASP.

Zu Tao¹, Sun Yi², Wu Dan¹, and Zhang Yi^1
1Key Laboratory for Biomedical Engineering of Ministry of Education, Department of Biomedical Engineering, College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou, China, ²MR Collaboration, Siemens Healthcare Ltd., Shanghai, China

The clinical use of chemical exchange saturation transfer (CEST) imaging is limited by its relatively long scan time, because it typically acquires multiple saturation image frames. Here, a novel auto-calibrated reconstruction method by joint k-space and image-space parallel imaging (KIPI) is proposed for faster CEST acquisition. By under-sampling CEST image frames with variable acceleration factors, KIPI allows an acceleration factor of up to 8-fold for acquiring source images, yielding a net speed-up of 6-fold in scan time, and produces image quality close to that of the ground truth.

Alexandra Grace Roberts^1,2, Pascal Spincemaille², Thanh Nguyen², and Yi Wang^2,3
1Electrical and Computer Engineering, Cornell University, Ithaca, NY, United States, ²Radiology, Weill Cornell, New York, NY, United States, ³Biomedical Engineering, Cornell University, Ithaca, NY, United States

Morphology Enabled Dipole Inversion (MEDI) is an iterative reconstruction algorithm for Quantitative Susceptibility Mapping (QSM) that is effective in suppressing streaking artifacts by exploiting the magnitude image as a morphological prior. However, contiguous areas of dipole incompatibility (such as noise) induce shadow artifacts whose spatial frequency components are not adequately regularized by the gradient based regularization in MEDI.  Here, we show the feasibility of adding a downsampled morphological prior to suppress these shadow artifacts.

Jan Martin¹, Patrik Brynolfsson^2,3, Michael Uder⁴, Frederik Bernd Laun⁴, Daniel Topgaard¹, and Alexis Reymbaut^2
1Physical Chemistry, Lund University, Lund, Sweden, ²Random Walk Imaging AB, Lund, Sweden, ³NONPI Medical AB, Umeå, Sweden, ⁴Institute of Radiology, University Hospital Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, Germany

Diffusion tensor distribution imaging (DTD) is a versatile technique enabling to retrieve nonparametric intra-voxel diffusion tensor distributions from tensor-valued diffusion-encoded data. While DTD owes its versatility to the minimal set of assumptions on which it relies, such minimal constraints induce a high sensitivity to noise hindering DTD's potential clinical translation. In this work, we demonstrate within a brain-like numerical phantom that generalized singular-value shrinkage (GSVS) denoising of the data prior to DTD analysis drastically improves DTD's accuracy, mitigating the aforementioned issue.

Hirohito Kan^1,2, Yuto Uchida³, Yoshino Ueki⁴, Satoshi Tsubokura⁵, Hiroshi Kunitomo⁵, Harumasa Kasai⁵, Noriyuki Matsukawa³, and Yuta Shibamoto^2
1Department of Integrated Health Sciences, Nagoya University Graduate School of Medicine, Nagoya, Japan, ²Department of Radiology, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan, ³Department of Neurology, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan, ⁴Department of Rehabilitation Medicine, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan, ⁵Department of Radiology, Nagoya City University Hospital, Nagoya, Japan

This study confirmed the feasibility of the simultaneous acquisition and voxel-based myelin water fraction (MWF), quantitative susceptibility mapping (QSM), and morphometry (VBM) analysis using magnetization-prepared multiple spoiled gradient echo (MP-mSPGR) sequence throughout comparisons with patients with Alzheimer’s disease and healthy control. As a result, the voxel-based MWF could detect demyelination in patients with AD. In contrast, there was not a significant change in susceptibility. This result suggested that the MWF depended on only the myelin content. The VBM analysis delineated the atrophy pattern of AD. The MP-mSPGR sequence is feasible for simultaneous MWF, QSM, and VBM analyses.

Nahla M H Elsaid¹, Nadine L Dispenza², R Todd Constable^1,3, Hemant D Tagare^1,2, and Gigi Galiana^1
1Radiology and Biomedical Imaging, Yale School of Medicine, New Haven, CT, United States, ²Department of Biomedical Engineering, Yale University, New Haven, CT, United States, ³Neurosurgery, Yale University, New Haven, CT, United States

Constrained alternating minimization for parameter mapping (CAMP) improves the image quality in highly accelerated parameter mapping by incorporating a linear constraint that relates consecutive images.

 

Samuel St-Jean^1,2, Filip Szczepankiewicz², Christian Beaulieu¹, and Markus Nilsson^2
1University of Alberta, Edmonton, AB, Canada, ²Clinical Sciences Lund, Lund University, Lund, Sweden

Parameter degeneracies and numerical fitting issues are the bane of diffusion MRI modelling. In this work, we present a framework using convex optimization to fit diffusion MRI models efficiently similarly to magnetic resonance fingerprinting. We also show how the singular value decomposition of these models can help to visualize how well the parameter space of a given model is sampled by a given acquisition protocol. Results on b-tensor encoded datasets and datasets leveraging multiple echo times literally show how these additional measurement dimensions disentangle model parameters better compared to traditional sequences.

Yiyun Dong¹, Qing-San Xiang^2,3, and Michael Nicholas Hoff^4
1Physics, University of Washington, Seattle, WA, United States, ²Physics, University of British Columbia, Vancouver, BC, Canada, ³Radiology, University of British Columbia, Vancouver, BC, Canada, ⁴Radiology, University of Washington, Seattle, WA, United States

bSSFP imaging enjoys widespread clinical use due to its high time efficiency and fluid-tissue contrast. Previous work with the elliptic signal model (ESM) and geometric solution (GS) inspire further analytical parameter quantification. Here the first exact solution for relaxation parameter T2 mapping is proposed based on the geometric properties of the ESM. Coupled with artifact-free images and field maps, a comprehensive imaging resource is generated. The analytic solution for source ESM parameters is proven to be relatively robust to variations in noise levels, T2 values and band proximity, and hence inspires a novel solution for T2 computation.

Nahla M H Elsaid¹, Nadine L Dispenza², R Todd Constable^1,3, Hemant D Tagare^1,2, and Gigi Galiana^1
1Radiology and Biomedical Imaging, Yale School of Medicine, New Haven, CT, United States, ²Department of Biomedical Engineering, Yale University, New Haven, CT, United States, ³Neurosurgery, Yale University, New Haven, CT, United States

This work presents quantitative T2-maps and a full T2w-image series generated from an ordinary single contrast T2w-dataset using the growing Constrained Alternating Minimization for Parameter mapping (g-CAMP) reconstruction method. Simulated data were used to study the accuracy of this approach under various echo spacings and with added noise, and the method is also demonstrated in an experimental T2w-dataset.  This could ultimately lead to retrospective parameter mapping using data from standard single-contrast acquisitions.

Yongquan Ye¹, Jingyuan Lv¹, Yichen Hu¹, Zhongqi Zhang², Jian Xu¹, and Weiguo Zhang^1
1UIH America, Inc., Houston, TX, United States, ²United Imaging Healthcare, Shanghai, China

In this work, we propose a robust and effective method of an accentuated T₁ weighted (aT₁W) method. By dividing T₁W images by proton density weighted (PDW) images, both obtained using GRE but with different flip angles, the resultant aT₁W images offer improved level of T₁ contrast. By adopting the multi-dimensional integration (MDI) technique, aT₁W images have high signal-to-noise ratio (SNR) despite the use of division operation.

Fan Yang¹, Jian Zhang², Guobin Li², Jiayu Zhu², Xin Tang¹, and Chenxi Hu^1
1Institute of Medical Imaging Technology, School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, ²United Imaging Healthcare Co., Ltd, Shanghai, China

Three-dimensional variable-flip-angle (VFA) T1 mapping is a valuable T1 quantification method subject to a long scan time due to acquisition of multiple 3D images. SUPER (Shift Undersampling improves Parametric-mapping Efficiency and Resolution) is a recently developed method providing fast blockwise reconstruction for accelerated parametric mapping. Here we develop a combination of SUPER and 3D CAIPIRINHA to achieve 5-fold acceleration with 5 flip angles, validated with both retrospective and prospective reconstruction. The results suggest that the proposed method is highly accurate for accelerating 3D VFA T1 mapping, reducing the whole-brain scan time from 6 to 1.5 minutes.

Gabrielle Grenier¹, François Rheault^1,2, and Maxime Descoteaux^1
1Sherbrooke Connectivity Imaging Lab (SCIL), Université de Sherbrooke, Sherbrooke, QC, Canada, ²Electrical Engineering, Vanderbilt University, Nashville, TN, United States

The Ensemble Average Propagator (EAP) encapsulates the angular and radial information of the diffusion magnetic resonance imaging (MRI) process and can highlight complex characteristics of the underlying tissue microstructure. Here, by sampling EAP values along different radii of the 3-way crossing of the centrum semiovale, we discriminate the orientations of the corticospinal tract from the arcuate fasciculus and the corpus callosum. This process finds microstructure signatures of fiber elements (fixels) of interest. Better tractometry as well as better local direction selection in tractography algorithms could be achieved with this information.

Venkata Veerendranadh Chebrolu¹, Xiaodong Zhong², Patrick Liebig³, and Robin Heidemann^3
1Siemens Medical Solutions USA, Inc., Rochester, MN, United States, ²Siemens Medical Solutions USA, Inc., Los Angeles, CA, United States, ³Siemens Healthcare GmbH, Erlangen, Germany

Uniform combined reconstruction (UNICORN) was recently proposed and used to improve receive uniformity of 7T musculoskeletal and brain MRI data from two-dimensional (2D) sequences. In this work, we extend the UNICORN algorithm to improve receive uniformity of two-, three-, or more-dimensional (N-dimensional or ND) MRI. We also demonstrate UNICORN results using L1-based (computationally efficient compared to singular value decomposition or SVD) optimal combination of the multi-channel data and compare the results with the previously used SVD-based combination.

Magnus Herberthson¹, Tom Dela Haije², Deneb Boito^3,4, Aasa Feragen⁵, Carl-Fredrik Westin⁶, and Evren Özarslan^3,4
1Dept. of Mathematics, Linköping University, Linköping, Sweden, ²Dept.of Computer Science, University of Copenhagen, Copenhagen, Denmark, ³Dept. of Biomedical Engineering, Linköping University, Linköping, Sweden, ⁴Center for Medical Image Science and Visualization, Linköping University, Linköping, Sweden, ⁵Department of Applied Mathematics and Computer Science, Technical University of Denmark, Lyngby, Denmark, ⁶Laboratory for Mathematics in Imaging, Department of Radiology, Brigham and Women’s Hospital, Harvard Medical School, Boston, Boston, MA, United States

Q-space trajectory imaging (QTI) provides a means to estimate the moments for a diffusion tensor distribution (DTD) characterizing the tissue. Commonly employed estimation methods do not typically yield mathematically acceptable estimates, namely that the tensor distribution consists of symmetric positive semidefinite tensors and that the second cumulant represents a covariance.  We introduce the QTI+ framework, which utilizes semi-definite programming (SDP) to address this issue. Simulations using a DTD taking the form of a non-central Wishart distribution as well as real data show a marked improvement in the technique's robustness to noise compared to more common (unconstrained) estimation methods.

Duohua Sun¹, Jean-Philippe Galons², Chidi Ugonna¹, Silu Han¹, and Nan-kuei Chen^1
1Biomedical Engineering, The University of Arizona, Tucson, AZ, United States, ²Medical Imaging, The University of Arizona, Tucson, AZ, United States

We present an approach for improving the phase variation representation of complex-valued dynamic MRI temperature mapping. Our technique utilizes phase information to better recover signal loss caused by susceptibility gradients and generate finer representations of dynamic phases variation. Results from numerical and hybrid simulation show that promising improvements in image resolution, susceptibility artifact reduction and phase variation representation can be achieved using our complex-valued super-resolution MRI scheme.

Zhuo-Xu Cui¹, Yuanyuan Liu², Qingyong Zhu¹, Jing Cheng², and Dong Liang^1,2
1Research center for Medical AI, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China, ²Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China

Thanks to powerful clinical applications, magnetic resonance (MR) parameter mapping has received widespread attentions. This work shows that the physical decay about parameter maps can be implicitly absorbed into an annihilation relation from k-space measurement of weighted MR images. Routinely, this relation can be estimated via null-space decomposition of a structured matrix, but, which usually results in computational burden. To alleviate it, we propose to train a convolutional neural network (CNN) to estimate this annihilation relation from undersampled measurement to further realize k-space interpolation. Experiments reveal the effectiveness of the proposed method compared with other competing methods.

Maarten Naeyaert¹, Vladimir Golkov², Daniel Cremers², Jan Sijbers³, and Marleen Verhoye^4
1Radiology, Universitair Ziekenhuis Brussel, Brussels, Belgium, ²Department of Computer Science, Technical University of Munich, Garching, Germany, ³Imec-Vision Lab, University of Antwerp, Wilrijk, Belgium, ⁴Bio-Imaging Lab, University of Antwerp, Wilrijk, Belgium

To accelerate the acquisition of HARDI data, compressed sensing can be used to subsample the data both in k-space and in q-space, using a holistic algorithm for the combined reconstruction. Fast spin echo (FSE) data has fewer deformation artefacts as EPI data, but often requires a multishot acquisition, making subsampling k-space more attractive. In this work FSE data was subsampled retrospectively to investigate different types of subsampling: subsampling q-space only, also using 1D k-space subsampling, or using q-space and alternated 1D k-space subsampling. The results show that for a given subsampling factor the alternated 1D k-space subsampling strategy performs best.

Jaeyong Yu^1,2, Sugil Kim³, Jae-Kyun Ryu⁴, and Jang-Yeon Park^1,2,4
1Department of Biomedical Engineering, Sungkyunkwan University, Suwon, Korea, Republic of, ²Department of Intelligent Precision Healthcare Convergence, Sungkyunkwan University, Suwon, Korea, Republic of, ³Siemens Healthineers, Seoul, Korea, Republic of, ⁴Biomedical Institute for Convergence at SKKU, Sungkyunkwan University, Suwon, Korea, Republic of

Acceleration imaging techniques such as parallel imaging and simultaneous multi-slice (SMS) imaging have been developed to achieve shorter acquisition time and larger FOV and incorporated into various imaging techniques including ultrafast imaging techniques such as echo-planar imaging (EPI). Here, we propose a new way to accelerate 3D spatiotemporal encoding (SPEN) imaging by combining SMS with controlled aliasing for parallel imaging results in higher acceleration (CAIPIRINHA) and split slice generalized auto-calibrating partially parallel acquisitions (Split Slice-GRAPPA).

Daiki Tamada¹ and Scott B. Reeder^1,2,3,4,5
1Radiology, University of Wisconsin-Madison, Madison, WI, United States, ²Medical Physics, University of Wisconsin-Madison, Madison, WI, United States, ³Biomedical Engineering, University of Wisconsin-Madison, Madison, WI, United States, ⁴Medicine, University of Wisconsin-Madison, Madison, WI, United States, ⁵Emergency Medicine, University of Wisconsin-Madison, Madison, WI, United States

A novel method for simultaneous T1 and T2-weighted imaging using RF phase-modulated GRE with small RF phase increments is presented. Configuration theory approach was used to derive an equation for the steady-state GRE signal. The equation reveals that small RF phase increments provide separable T1 and T2-weighted contrast in the real and imaginary components of the signal. Simulation and phantom studies were performed for quantitative analysis of the proposed method. Brain in-vivo imaging was included to show clinical feasibility. The results suggested the proposed method enables faster imaging compared to conventional FSE imaging.