Rong Guo^1,2, Yudu Li^1,2, Yibo Zhao^1,2, Tianyao Wang³, Yao Li⁴, Brad Sutton^1,2,5, and Zhi-Pei Liang^1,2
1Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ²Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ³Radiology Department, Fifth People's Hospital of Shanghai, Fudan University, Shanghai, China, ⁴School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, ⁵Department of Bioengineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States

The feasibility of high-resolution metabolic imaging using water-unsuppressed MRSI has been recently demonstrated in clinical settings using SPICE. This work utilizes the unsuppressed water spectroscopic signals for distortion-free mapping of water diffusion coefficients, thus adding a new feature to SPICE for multi-modal brain mapping. Experimental results demonstrated that simultaneous distortion-free diffusion imaging at 1.0×1.0×2.0 mm³ resolution and metabolic imaging at 2.0×3.0×3.0 mm³ nominal resolution were successfully obtained in a total 8-min scan.

Lipeng Ning¹ and Yogesh Rathi^1
1Harvard Medical School, BWH, Boston, MA, United States

Diffusion MRI is sensitive to heterogeneity tissue microstructure. Recent methods for axon diameter estimation rely on diffusion MRI data acquired using advanced scanners with b-values higher than 6000 s/mm2 so that signals from extra-axonal components vanish because of the underlying relative high diffusivity. In this work, we introduce a method for axon diameter estimation that uses diffusion MRI with two TEs and relative lower b-values acquired using clinical scanners. Our method separates signals from intra- and extra-axonal components by exploiting the b-value-dependent relaxation coefficients. The performance of the proposed method is demonstrated using in vivo data acquired from a clinical scanner.

Nastaren Abad¹, Radhika Madhavan¹, Tim Sprenger², Chitresh Bhushan¹, Ante Zhu¹, Luca Marinelli¹, Eric Fiveland¹, Seung-Kyun Lee¹, J. Kevin DeMarco³, Robert Shih³, Maureen Hood^3,4, Gail Kohls³, H Doug Morris³, Kimbra Kenny⁴, Vincent B Ho^3,4, and Thomas KF Foo^1
1GE Global Research Center, Niskayuna, NY, United States, ²MR Applied Science Laboratory Europe, GE Healthcare, Stockholm, Sweden, ³Walter Reed National Military Medical Center, Bethesda, MD, United States, ⁴Uniformed Services University of the Health Sciences, Bethesda, MD, United States

An ultra-high diffusion b-value protocol (b=7,000-30,000s/mm²) was acquired in a high-performance 3.0 T MAGNUS head gradient system (max gradient amplitude=200 mT/m and max slew rate=500 T/m/s) for white matter microstructure imaging.With shorter TE enabled by MAGNUS, measurements of intra-axonal diffusivities are demonstrated at ultra-high b-values. At these ultra-high b-values, signal from extra-axonal water is effectively suppressed but SNR is also reduced. We show that real-valued image reconstruction maintains Gaussian noise properties and reduces Rician bias, allowing better quantification of intra-axonal diffusivities and effective axonal radii in healthy volunteers and in ongoing human subject research studies of mild traumatic brain injury.

Philip Kenneth Lee^1,2, Daehyun Yoon², and Brian Andrew Hargreaves^1,2,3
1Electrical Engineering, Stanford University, Stanford, CA, United States, ²Radiology, Stanford University, Stanford, CA, United States, ³Bioengineering, Stanford University, Stanford, CA, United States

Metallic implants create severe off-resonance adjacent to the implant, in the range of 2-20 kHz at 3T. The off-resonance pattern creates a static, “always on” background gradient with unknown amplitude and polarity. Static non-linear background gradients are a known source of local encoding errors, but their effect on diffusion contrast near implants has not yet been evaluated. Static background gradients are distinct from diffusion gradient non-linearities, which affect the achieved b-value if sufficiently large. We apply theoretical results and phantom experiments to evaluate two different diffusion encoding schemes and evaluate the effect of metal-induced off-resonance on ADC quantification.

Mark Barahman¹, Eduardo Grunvald², Pablo J Prado³, Alejandro Bussandri³, Walter C Henderson¹, Tanya Wolfson¹, Kathryn J Fowler¹, and Claude B Sirlin^1
1Radiology, UCSD, San Diego, CA, United States, ²Internal Medicine, UCSD, San Diego, CA, United States, ³Livivos Inc, San Diego, CA, United States

LiverScope® is a novel point-of-care (POC) NMR technology for liver fat quantification. 

·       POC NMR PDFF agreed closely with ground-truth PDFF values in commercial PDFF phantoms (R² = 0.99)

·       POC NMR was feasible in an outpatient clinic, had no demonstrated adverse events, and agreed closely with contemporaneous MRI-PDFF reference standard values (R² = 0.94) in human adults with obesity and at risk for NAFLD: 

Jordan A. Chad^1,2, Nir Sochen^3,4, J. Jean Chen^1,2, and Ofer Pasternak^5,6
1Rotman Research Institute, Baycrest Health Sciences, Toronto, ON, Canada, ²Department of Medical Biophysics, University of Toronto, Toronto, ON, Canada, ³School of Mathematical Sciences, Tel Aviv University, Tel Aviv, Israel, ⁴School of Neuroscience, Tel Aviv University, Tel Aviv, Israel, ⁵Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, United States, ⁶Department of Psychiatry, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, United States

Recent diffusion modeling studies aim to distinguish age-related degenerative processes specific to white matter fibers, such as axonal myelination and dispersion, from age-related elevated isotropic diffusivity, resulting from, e.g., enlargement of the extracellular space. Here we analytically investigate the relationship between two single-shell approaches for distinguishing these effects, respectively based on a one- and two-compartment model. We derive a nonlinear relationship between the models that renders a sensitivity advantage to the two-compartment model, and find that a linear elevation of an isotropic compartment with age explains the well-documented phenomenon of accelerated MD elevations (faster radially than axially) in aging.

Patryk Filipiak¹, Lee Basler², Benjamin Rodack², John Dzikiy², Anthony Zuccolotto², Ying-Chia Lin¹, Dimitris G. Placantonakis³, Timothy Shepherd¹, Walter Schneider⁴, Fernando E. Boada¹, and Steven H. Baete^1
1Center for Advanced Imaging Innovation and Research (CAI2R), Department of Radiology, NYU Langone Health, New York, NY, United States, ²Psychology Software Tools, Inc., Pittsburgh, PA, United States, ³Department of Neurosurgery, Perlmutter Cancer Center, Neuroscience Institute, Kimmel Center for Stem Cell Biology, NYU Langone Health, New York, NY, United States, ⁴University of Pittsburgh, Pittsburgh, PA, United States

We optimize the sampling length parameter of Radial Diffusion Spectrum Imaging (RDSI) to improve fiber reconstruction accuracy of ODF-Fingerprinting (ODF-FP). As the ground truth, we use anisotropic diffusion phantoms containing taxons (textile water-filled tubes) with inside diameter of 0.8µm, approaching the anatomical scale of axons found in the human brain. Thanks to our optimized approach, ODF-FP reconstructs fibers crossing at angles as narrow as 10, 15, and 20 degrees with the average accuracy of approximately 10 degrees. We generalize our observations by proposing a criterion for selecting near-optimal sampling lengths to accelerate and simplify the application of ODF-FP with RDSI.
 

Erick Hernandez-Gutierrez¹, Alonso Ramirez-Manzanares², Antoine Théberge¹, Maxime Descoteaux¹, Luis Concha³, and Ricardo Coronado-Leija^4
1Université de Sherbrooke, Sherbrooke, QC, Canada, ²Centro de Investigación en Matemáticas A.C., Guanajuato, Mexico, ³Universidad Nacional Autónoma de México, Querétaro, Mexico, ⁴New York University School of Medicine, New York, NY, United States

In this study, we propose a pipeline to compute the structural connectome from diffusion-weighted magnetic resonance imaging (dMRI). The pipeline is based on a novel method called MRDS, which computes independent Gaussian profiles per intravoxel diffusion compartments. The effectiveness of our pipeline is shown on the DiSCo challenge synthetic dataset. Our findings show that the pipeline is competitive and outperforms a pipeline based on constrained spherical deconvolution (CSD).

Charles Poirier¹ and Maxime Descoteaux^1
1Computer Sciences, Université de Sherbrooke, Sherbrooke, QC, Canada

Although the voxel-wise signal acquired by diffusion MRI is symmetric, it is not always the case of the underlying fiber configurations. We present angle-aware bilateral filtering, an intuitive method for denoising fiber ODF fields revealing asymmetric fiber ODFs. To evaluate the effect of the filtering, tractography is performed on the resulting asymmetric fiber ODF field. Compared to the tractogram obtained from the original fiber ODF field, we show that using asymmetric filtered fiber ODF field as input to a fiber tracking algorithm reduces the ratio of incomplete streamlines and false connections and increases the proportion of true connections.

Daniel Djayakarsana¹, Greg Stanisz¹, and Colleen Bailey^1
1Sunnybrook Research Institute, Toronto, ON, Canada

Modeling diffusion that incorporates the microscopic partially absorbing wall is a more realistic description compared to the standard Bloch equation based two-site model with exchange. The resulting signal equations are more complex and involve working in the Laplace domain. Here, we demonstrate that the increased spatial relevance of the partially absorbing wall model does not improve fitting in our AML model treated with cisplatin for cell death. In fact, the partially absorbing wall model fits and parameters are poorer. Therefore, extracting diffusion parameters from the Bloch equation based two-site exchange model is sufficient for our diffusion MRI datasets.

Sean P Devan^1,2, Xiaoyu Jiang^1,3, Kurt G Schilling^1,3,4, Feng Wang^1,3, Li Min Chen^1,3, John C Gore^1,3,4,5, and Junzhong Xu^1,3,4,5
1Vanderbilt University Institute of Imaging Science, Vanderbilt University Medical Center, Nashville, TN, United States, ²Chemical and Physical Biology Program, Vanderbilt University, Nashville, TN, United States, ³Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, United States, ⁴Department of Biomedical Engineering, Vanderbilt University, Nashville, TN, United States, ⁵Department of Physics and Astronomy, Vanderbilt University, Nashville, TN, United States

Time-dependent diffusivity is sensitive to microstructural geometry and has previously been described by a power-law scaling. However, the limits of validity of this relationship across different ranges of diffusion time are unclear.  We therefore acquired brain images of nonhuman primates in vivo with diffusion times from 1.9 to 40 ms to assess how well a power-law fits diffusivity in this range. We assess fitting in the time and frequency domains, whether pulsed (PGSE) and oscillating (OGSE) gradient spin echo can be combined for better results, and the sensitivity of diffusivity estimates to varying power-law exponents. 

Priya S Balasubramanian^1,2, Lingfei Guo^2,3, Weiyuan Huang^2,4, Ilhami Kovanlikaya², Alisa Ramineni⁵, Nector Garcia⁵, Benjamin Liechty⁵, Thanh Nguyen², Pascal Spincemaille², David Pisapia⁵, and Yi Wang^2
1Electrical and Computer Engineering, CORNELL UNIVERSITY, New York City, NY, United States, ²Department of Radiology, Weill Cornell Medicine, New York City, NY, United States, ³Department of Radiology, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan, Shandong, Jinan, Shandong, China, ⁴Department of Radiology, Hainan General Hospital, Hainan Hospital Affiliated to Hainan Medical College, Hainan, China, ⁵Department of Pathology and Laboratory Medicine, Weill Cornell Medicine and New York Presbyterian Hospital, New York City, NY, United States

Quantitative susceptibility mapping (QSM) solves an ill-posed dipole field inversion problem and is prone to streaking and shadowing artifacts. This report presents a novel QSM reconstruction algorithm that effectively suppresses these artifacts using prior information from traditional MRI contrasts. The proposed algorithm was evaluated using ex vivo high resolution QSM of cerebellum and brainstem specimens using COSMOS as the reference method, showing improved QSM accuracy and artifact reduction and better agreement with myelin sensitive Luxol Fast Blue staining compared to conventional MEDI and MEDI-SMV methods.  

Javad Hamidi Esfahani¹, Nashwan Naji¹, Peter Seres¹, Christian Beaulieu¹, and Alan H Wilman^1
1Biomedical Engineering, University of Alberta, Edmonton, AB, Canada

Susceptibility separation enables a voxel-wise decomposition into paramagnetic and diamagnetic components. The repeatability of the method is examined in four structures of deep gray matter and three of white matter using scan rescan data of 22 subjects. The paramagnetic susceptibility separation is shown to be more robust than the diamagnetic and total susceptibility is shown to be similar to conventional QSM. Thus, susceptibility separation can provide both qualitative depiction of diamagnetic and paramagnetic susceptibility components and repeatable measures of paramagnetic components at least in deep grey matter.

Priya S Balasubramanian^1,2, Lingfei Guo^2,3, Weiyuan Huang^2,4, Ilhami Kovanlikaya², Alisa Ramineni⁵, Nector Garcia⁵, Benjamin Liechty⁵, Jonathan Dyke², Henning Voss², Eric Aronowitz², Pascal Spincemaille², David Pisapia⁵, and Yi Wang^2
1Electrical and Computer Engineering, CORNELL UNIVERSITY, New York City, NY, United States, ²Department of Radiology, Weill Cornell Medicine, New York City, NY, United States, ³Department of Radiology, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan, Shandong, Jinan, Shandong, China, ⁴Department of Radiology, Hainan General Hospital, Hainan Hospital Affiliated to Hainan Medical College, Hainan, China, ⁵Department of Pathology and Laboratory Medicine, Weill Cornell Medicine and New York Presbyterian Hospital, New York City, NY, United States

This study utilized sub-millimeter 3T gradient echo MRI of the brainstem ex vivo to obtain a magnetic susceptibility map (QSM). ROIs of both deep brain nuclei and myelinated fiber tracts were chosen and quantified from QSM obtained from MRI field. The ex vivo specimens were then stained with Luxol Fast Blue and the resultant absorbance of the ROIs were compared to QSM. A strong linear correlation was observed (R² = 0.897, p = 0.0041). 

Jinwei Zhang¹, Alexey Dimov², Hang Zhang¹, Chao Li¹, Thanh Nguyen², Pascal Spincemaille², and Yi Wang^2
1Cornell University, New York, NY, United States, ²Weill Cornell Medicine, New York, NY, United States

A network fine-tuning step based on signal physics is proposed for deep learning based quantitative susceptibility mapping using high-pass filtered phase only to susceptibility. The proposed method showed better robustness compared to the pre-trained networks without fine-tuning when the test dataset deviated from the training dataset, such as a change in voxel size or high-pass filter cutoff frequency.

Nashwan Naji¹ and Alan H. Wilman^1
1Department of Biomedical Engineering, University of Alberta, Edmonton, AB, Canada

Susceptibility maps reconstructed from small-axial slabs suffer underestimation due to background-field removal imperfections near slab boundaries and the increased difficulty of solving a 3D-inversion problem with reduced-support in the direction of B0. Reliable QSM reconstruction from small slabs would enable higher resolution imaging within a reasonable time-frame and further accelerate clinical adoption. This work proposed using additional rapid low-resolution data, that serves as prior information, to improve background-field estimation and regularize the inversion-to-susceptibility process. Consistent high-resolution QSM at 3T was obtained from slabs of width as small as 6.8mm, aided by a lower-resolution dataset of 16 times coarser voxels.