View Presentation Video

Fábio Seiji Otsuka¹, Maria Concepción García Otaduy², and Carlos Ernesto Garrido Salmon^1
1InbrainLab, University of São Paulo, Ribeirão Preto, Brazil, ²LIM44, University of São Paulo, São Paulo, Brazil

Keywords: Electromagnetic Tissue Properties, Quantitative Susceptibility mapping

QSM is a promising MRI techniques that enables the assessment of magnetic properties of tissue. However, its underlying biophyisical source of contrast is still unknown. It has been shown that iron is well correlated in the basal ganglia, however this hasn't been investigates for other regions. Furthermore, iron can be presented in different forms in the brain, each one having different functions and properties. This work aims to assess the composition of brain structures regarding its paramagnetic ions' and total metal content, comparing their concentrations with magnetic susceptibility assessed by QSM

View Presentation Video

Sooyeon Ji¹, Juhyung Park¹, Hyeong-Geol Shin^2,3, Joonhyeok Yoon¹, Minjun Kim¹, and Jongho Lee^1
1Department of Electrical Computer Engineering, Seoul National University, Seoul, Korea, Republic of, ²Department of Radiology, Johns Hopkins University School of Medicine, Baltimore, MD, United States, ³F.M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States

Keywords: Susceptibility, Data Processing

A pipeline to reconstruct multiple resolution QSM data using a QSM network trained at a single resolution is proposed. The local field map is re-sampled multiple times in different spatial locations, and the re-sampled local field maps are used to reconstruct QSM maps at training data resolution. The reconstructed maps are then combined, and corrected for using a procedure named “dipole compensation”. When compared to two scenarios to reconstruct different resolution data using network trained at a single resolution, the proposed pipeline demonstrated the best performance both qualitatively and quantitatively.

View Presentation Video

Hirohito Kan¹, Yuto Uchida², Yoshino Ueki³, and Harumasa Kasai^4
1Department of Integrated Health Scieneces, Nagoya University Graduate School of Medicine, Nagoya, Japan, ²Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, United States, ³Department of Rehabilitation Medicine, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan, ⁴Department of Radiology, Nagoya City University Hospital, Nagoya, Japan

Keywords: Electromagnetic Tissue Properties, White Matter

The magnetic source separation method can estimate contributions of diamagnetic myelin and paramagnetic iron in the white matter using solely gradient-echo data in white matter. This study determined the relationship between negative susceptibility (χ-) and fractional anisotropy as a myelin-sensitive biomarker and compared with the conventional susceptibility (χ) in young healthy volunteers. There was a significant negative correlation between the χ- and FA. In contrast, the conventional χ has a weaker correlation to the FA than the χ- result. However, the χ- in white matter represented the non-monotonic fiber orientation dependence on the B0 field.

View Presentation Video

Julia V Velikina¹, Collin J Buelo^1,2, Yan Wu³, Marcus T Alley³, Moniba Nazeef⁴, Michael Jeng³, Alexey A Samsonov¹, Scott B Reeder^1,2,4,5,6, Shreyas S Vasanawala³, and Diego Hernando^1,2
1Radiology, University of Wisconsin - Madison, Madison, WI, United States, ²Medical Physics, University of Wisconsin - Madison, Madison, WI, United States, ³Stanford University, Stanford, CA, United States, ⁴Medicine, University of Wisconsin - Madison, Madison, WI, United States, ⁵Biomedical Engineering, University of Wisconsin - Madison, Madison, WI, United States, ⁶Emergency Medicine, University of Wisconsin - Madison, Madison, WI, United States

Keywords: Susceptibility, Liver

Quantitative susceptibility mapping (QSM) is a promising non-invasive technique for quantification of liver iron concentration. Abdominal QSM typically requires a breath-hold acquisition since respiration induces liver motion, which leads to blurring artifacts. However, some patients have trouble even with a short breath-hold, which necessitates development of free-breathing approaches. In this work, we report initial results on the feasibility of using the modified “butterfly” navigator approach in multi-echo imaging in conjunction with compressed sensing reconstruction to enable free-breathing liver QSM.

Mathias Lambert^1,2,3, Javier Silva^1,2,3, Carlos Milovic⁴, and Cristian Tejos^1,2,3
1Department of Electrical Engineering, Pontificia Universidad Católica de Chile, Santiago, Chile, ²Biomedical Imaging Center, Pontificia Universidad Católica de Chile, Santiago, Chile, ³Millennium Institute for Intelligent Healthcare Engineering (iHEALTH), Santiago, Chile, ⁴School of Electrical Engineering, Pontificia Universidad Católica de Valpaiso, Valparaiso, Chile

Keywords: Susceptibility, Quantitative Susceptibility mapping

Guiding the network architecture to learn to apply the kernel inversely in Fourier space allows training to be less prone to overfitting. Using simulated images from a single brain image, it is possible to satisfactorily reconstruct a susceptibility map of the abdomen. By having as input the Fourier space of the local field and the kernel of the dipole, the network learned to reduce the noise, to divide the data of the local field by the kernel where possible and to recover the data in the magic cone. 

 

View Presentation Video

Elena Grosso¹, Antonio Ricciardi², Egidio D'Angelo^1,3, Fulvia Palesi¹, and Claudia AM Gandini Wheeler-Kingshott^1,2,3
1Department of Brain and Behavioral Sciences, University of Pavia, Pavia, Italy, ²NMR Research Unit, Department of Neuroinflammation, Queen Square Multiple Sclerosis Centre, UCL Queen Square Institute of Neurology, University College London (UCL), London, United Kingdom, ³Brain Connectivity Centre Research Unit, IRCCS Mondino Foundation, Pavia, Italy

Keywords: Susceptibility, Modelling, Brain

χ-separation is a recently proposed biophysical model to separate the χ-positive and χ-negative contribution of magnetic susceptibility sources in the brain. Such a model relies on assuming a certain relaxometric constant (D_r) calculated as the mean of a group of healthy subjects.  Here we demonstrate that D_r is subject-specific and if used in place of the average value, it affects the χ-positive and  χ-negative maps in a structured way, while their sum remains similar. 

View Presentation Video

Oliver C. Kiersnowski¹, David L. Thomas², Adam K. Yamamoto^2,3, Mohammed Elgwely^2,3, Anastasia Papadaki^2,3, Tarek Yousry^2,3, John S. Thornton^2,3, and Karin Shmueli^1
1Department of Medical Physics and Biomedical Engineering, University College London, London, United Kingdom, ²Neuroradiological Academic Unit, UCL Queen Square Institute of Neurology, University College London, London, United Kingdom, ³Lysholm Department of Neuroradiology, National Hospital for Neurology and Neurosurgery, London, United Kingdom

Keywords: Susceptibility, Quantitative Susceptibility mapping

Quantitative susceptibility mapping (QSM) has been used to investigate movement disorders but has not been integrated into routine clinical practice. We developed an acquisition protocol and a robust QSM pipeline for neuroradiological investigation of movement disorders. We show that high quality QSMs can be acquired using a multi-echo 3D gradient-echo sequence with partial k-space filling in under 6 minutes with only one of eleven patient QSMs corrupted by motion artifacts. We show that Laplacian phase unwrapping and projection onto dipole fields (PDF) background field removal are robust to artifacts across patients with strong susceptibility sources associated with various pathologies.

View Presentation Video

Nestor Munoz^1,2,3, Carlos Milovic^3,4, Christian Langkammer⁵, Sergio Uribe^2,3,6, and Cristian Tejos^1,2,3
1Electrical Engineering, Pontificia Universidad Catolica de Chile, Santiago, Chile, ²Biomedical Imaging Center, Pontificia Universidad Catolica de Chile, Santiago, Chile, ³Millennium Institute for Intelligent Healthcare Engineering (iHEALTH), Santiago, Chile, ⁴School of Electrical Engineering, Pontificia Universidad Católica de Valparaiso, Valparaiso, Chile, ⁵Department of Neurology, Medical University of Graz, Graz, Austria, ⁶Department of Radiology, Pontificia Universidad Catolica de Chile, Santiago, Chile

Keywords: Susceptibility, Susceptibility

Diffusion Tensor Imaging (DTI) and Susceptibility Tensor Imaging (STI) are two MRI techniques that produces anisotropy information and fiber direction of biological tissue. For the brain, it has been suggested that both principal eigenvectors (PEV) point towards the same direction (i.e., direction of myelinated axons). However, different resolution of both techniques might produce differences in the PEVs. We proposed a quantitative method to compare DTI and STI PEVs based on the cosine of their angular difference. As expected, we found that PEVs from STI and DTI show similar directions in white matter and that consistency is lost for gray matter and CSF. 

View Presentation Video

Haodong Zhong¹, Yi Wang², and Jianqi Li^1
1Shanghai key lab of magnetic resonance, East China Normal University, Shanghai, China, ²Department of Radiology, Weill Medical College of Cornell University, New York, NY, United States

Keywords: Susceptibility, Quantitative Susceptibility mapping, Total Field Inversion

This study proposes a novel total field inversion method in which the background field is modeled by discrete boundary elements. In this method, the boundary value of background field and local tissue susceptibility are simultaneously estimated in one step. We validated our method with orthogonality numerical simulation and in vivo data.

View Presentation Video

Lin Chen^1,2, Anja Soldan³, Zixuan Lin¹, Kumiko Oishi⁴, Kenichi Oishi¹, Andreia Faria¹, Marilyn Albert³, Peter van Zijl^1,2, and Xu Li^1,2
1Department of Radiology and Radiological Sciences, Johns Hopkins University, Baltimore, MD, United States, ²F.M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States, ³Department of Neurology, Johns Hopkins University, Baltimore, MD, United States, ⁴Center for Imaging Science, Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD, United States

Keywords: Susceptibility, Aging

We investigated associations of brain iron as measured by QSM, global oxygen extraction fraction (OEF) as measured by TRUST and white matter hyperintensity (WMH) as measure by FLAIR, and their possible interactive effects on both global composite and domain-specific cognitive functions in cognitively normal participants. Significant associations were observed between global WMH burden and tissue susceptibility suggesting contributions from small vessel diseases to tissue iron deposition. Negative associations between tissue susceptibility and cognition as well as positive associations between OEF and cognitive performance were observed within participants with low WMH burden, but not as significant in high WMH group.

View Presentation Video

Anders Dyhr Sandgaard¹, Valerij G. Kiselev², Noam Shemesh³, and Sune Nørhøj Jespersen^1,4
1Center for functionally integrative neuroscience, department of clinical medicine, Aarhus University, Aarhus, Denmark, ²Division of Medical Physics, Department of Radiology, University Medical Center Freiburg, Freiburg, Germany, ³Champalimaud Research, Champalimaud Centre for the Unknown, Lisbon, Portugal, ⁴Department of Phsysics and Astronomy, Aarhus University, Aarhus, Denmark

Keywords: Microstructure, Quantitative Susceptibility mapping

Quantitative Susceptibility Mapping (QSM) is a highly utilized MRI modality for mapping tissue susceptibility. However, a limitation of QSM is disregarding mesoscopic field effects associated with WM microstructure and anisotropic susceptibility. Here we present a minimal extension of QSM by including frequency shifts due to the fibrous WM microstructure, while still neglecting susceptibility anisotropy and WM spherical inclusions modelling iron complexes. We find that this step already improves the accuracy of QSM as it is shown by comparison with conventional QSM using a digital phantom that includes microstructural frequency shifts from multiple sources.

View Presentation Video

Carlos Milovic^1,2, Patrick Fuchs³, Oriana Arsenov³, Oliver C Kiersnowski³, Russell Murdoch³, Laxmi Muralidharan³, Jannette Nassar³, and Karin Shmueli^3
1School of Electrical Engineering, Pontificia Universidad Catolica de Valparaiso, Valparaiso, Chile, ²iHEALTH, Millennium Institute for Intelligent Healthcare Engineering, Santiago, Chile, ³University College London, London, United Kingdom

Keywords: Susceptibility, Susceptibility, QSM

Removing background fields is an important preprocessing step in QSM, enabling the reconstruction of fine tissue susceptibility variations in the region of interest (ROI) without being corrupted by susceptibility sources outside of this region. This requires a binary mask of the ROI and most background field removal methods are sensitive to the choice of mask. Here we compared 15 background field removal methods across 4 different masks. We found projection onto dipole fields (PDF) to perform best overall, although it is sensitive to the mask. V-SHARP and RESHARP were more robust to masking and showed good performance.

View Presentation Video

Alexander Jaffray¹, Lara Bartels¹, Christian Kames¹, and Alexander Rauscher^1,2
1Physics, UBC MRI Research Centre, Vancouver, BC, Canada, ²Department of Pediatrics, UBC, Vancouver, BC, Canada

Keywords: Susceptibility, Brain, small vessel, central vein, phase, regularization, vein, qsm, quantitative, mapping

Quantitative susceptibility mapping is a possible modality with which to measure and identify diseases involving small vasculature in the brain. However, multi-echo gradient echo images that are a requirement for high quality quantitative susceptibility map reconstruction require long scan durations, limiting clinical utility. This duration can be reduced with the use of parallel imaging strategies, and non-linear reconstruction with total generalized variation regularization is an increasingly popular strategy for undersampled reconstruction. We demonstrate that the depiction of small vessels and surrounding structures in QSM is significantly sensitive to the choice of regularization parameter used in the reconstruction.

View Presentation Video

Ming Zhang¹, Yuyao Zhang², and Hongjiang Wei^1
1School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, ²School of Information and Science and Technology, ShanghaiTech University, Shanghai, China

Keywords: Susceptibility, Quantitative Susceptibility mapping, implicit neural representation

This study introduced an unsupervised deep learning-based method for QSM reconstruction using implicit neural representation (INR-QSM), a training databases-free method for high-quality QSM reconstruction. In INR-QSM, the susceptibility map was represented as a continuous function of the spatial coordinates. A coordinate-based multilayer perceptron (MLP) parameterized this function, took the coordinate as input and predicted the susceptibility value at the corresponding spatial location. The parameters of MLP were updated by minimizing a custom cost function. Preliminary results on two different datasets demonstrated the potential of INR for unsupervised QSM reconstruction.

View Presentation Video

Ming Zhang¹, Lanlan Gao², Shuheng Zhang², and Hongjiang Wei^1
1School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, ²United Imaging Healthcare, Shanghai, China

Keywords: Susceptibility, Quantitative Susceptibility mapping, cerebral cortex

It is challenging to resolve the human brain microstructure using MRI due to limited SNR and poor image contrast. High-resolution quantitative susceptibility mapping (QSM) at an ultra-high magnetic field could be a useful tool for revealing the subtle structure of the brain due to intrinsic susceptibility differences between tissues. In this study, the six-layer laminae of the cerebral cortex and swallow-tail sign of SN were successfully resolved using high-resolution QSM at 5.0-T. The preliminary result could indicate the potential of high-resolution QSM for investigating the subtle structures of the human brain and related brain disorders.

View Presentation Video

De-Rong Huang¹, Jhih-Shan Cheng¹, Hsiao-Wen Chung², and Ming-Long Wu^1,3
1Department of Computer Science and Information Engineering, National Cheng Kung University, Tainan, Taiwan, ²Graduate Institute of Biomedical Electronics and Bioinformatics, National Taiwan University, Taipei, Taiwan, ³Institute of Medical Informatics, National Cheng Kung University, Tainan, Taiwan

Keywords: Susceptibility, Machine Learning/Artificial Intelligence

A deep learning framework termed pCOSMOS for quantitative susceptibility mapping (QSM) was proposed, which employed a local-field-to-local-field transformation to generate multiple orientation local field maps from single orientation data, followed by COSMOS reconstruction using physical model.  3T multi-orientation data from 10 healthy subjects (7 for training, 3 for testing) were used to investigate performance.  Quantitative results compared with MEDI, SFCR, QSMnet, and LPCNN demonstrated superior performance of pCOSMOS comparable to the amongst best algorithms using single orientation data, while showing drastically reduced training time from tens of hours to 3.2 hours.

View Presentation Video

Jacob Degitz¹, Edith Valle², Steven M Wright², and Mary P McDougall^1
1Biomedical Engineering, Texas A&M University, College Station, TX, United States, ²Electrical Engineering, Texas A&M University, College Station, TX, United States

Keywords: Susceptibility, Non-Array RF Coils, Antennas & Waveguides

One of the earliest significant events in Duchenne muscular dystrophy (DMD) pathology is an elevated influx of calcium ions into afflicted cells. To study this event and its correlation with disease progression, a custom 22mm diameter quadrature volume coil was designed and used to obtain ¹H images at 4.7T, the processing protocol was established, and comparisons were made to μCT images.

View Presentation Video

Korbinian Eckstein¹, Thanh Thuy Dao², Ashley Stewart^2,3, Simon Daniel Robinson^4,5,6,7, Markus Barth^2,3,4, and Steffen Bollmann^2,3,4
1School of Information Technology and Electrical Engineering, The University of Queensland, Dutton Park, Australia, ²School of Information Technology and Electrical Engineering, The University of Queensland, Brisbane, Australia, ³ARC Training Centre for Innovation in Biomedical Imaging Technology, The University of Queensland, Brisbane, Australia, ⁴Centre for Advanced Imaging, The University of Queensland, Brisbane, Australia, ⁵Department of Neurology, Medical University of Graz, Graz, Austria, ⁶Karl Landsteiner Institute for Clinical Molecular MR in Musculoskeletal Imaging, Vienna, Austria, ⁷High Field MR Center, Medical University of Vienna, Vienna, Austria

Keywords: Electromagnetic Tissue Properties, Quantitative Susceptibility mapping, Multi-echo, ROMEO, Phase

With increasingly popular multi-echo QSM, the combination of echoes becomes important in terms of accuracy, SNR and computation time. We compared 6 different pipelines with quantitative and Laplacian unwrapping and three different echo combination approaches, weighted averaging of QSMs, combination of the phase, and non-linear fitting. We compared the pipelines on the QSM challenge brain dataset and 7 T in vivo data and conclude that quantitative unwrapping (ROMEO) with weighted frequency combination achieves the best outcomes in terms of accuracy, SNR and computation time.

View Presentation Video

Mónica Ferreira¹, Rüdiger Stirnberg², Yannik Völzke², Daniel Löwen², Christian Langkammer³, Simon Robinson^3,4, Stefan Ropele³, Thomas Klockgether^2,5, Tony Stöcker^2,6, and Jennifer Faber^2,5
1Clinical Research, DZNE Bonn, Bonn, Germany, ²DZNE Bonn, Bonn, Germany, ³Department of Neurology, Medical University of Graz, Graz, Austria, ⁴High Field MR Center, Department of Biomedical Imaging and Image-Guided Therapy, Medical University of Vienna, Vienna, Austria, ⁵Department of Neurology, University Hospital Bonn, Bonn, Germany, ⁶Department of Physics and Astronomy, University of Bonn, Bonn, Germany

Keywords: Susceptibility, Quantitative Susceptibility mapping

RF transmit field inhomogeneities at 7T usually degrade the image quality in regions such as the cerebellum and brainstem, which has so far delayed its use to study cerebellar ataxias with higher fields. We present an optimized pipeline that maximizes the QSM contrast of nuclei in the cerebellum based on multi-echo MPM data. We show that averaging the individual susceptibility maps increases the signal-to-noise ratio (SNR) and reduces the group standard deviation. We introduced a weighted averaging based on the individual acquisition SNRs. Final maps presented high contrast of the dentate nucleus and a high delineation of its denticulated silhouette. 

View Presentation Video

Oliver C. Kiersnowski¹, Patrick Fuchs¹, Stephen J. Wastling^2,3, John S. Thornton^2,3, and Karin Shmueli^1
1Department of Medical Physics and Biomedical Engineering, University College London, London, United Kingdom, ²Neuroradiological Academic Unit, UCL Queen Square Institute of Neurology, University College London, London, United Kingdom, ³Lysholm Department of Neuroradiology, National Hospital for Neurology and Neurosurgery, London, United Kingdom

Keywords: Susceptibility, Quantitative Susceptibility mapping

Echo planar imaging (EPI) suffers from geometric distortions, which can be corrected using reference acquisitions for single-echo EPI, or using field maps calculated from the multiple echoes in multi-echo EPI. The effect of distortion on EPI quantitative susceptibility mapping (QSM) has not been investigated. We compared static and dynamic geometric distortion correction for magnitude images and QSM using multi-echo 2D-EPI acquisitions with and without changes in head position. Multi-echo EPI corrected distortion in both magnitude and QSM images without the need for reference scans. Correcting the local field map in the QSM pipeline was optimal to reduce temporal variance.