Arthur Chakwizira¹, Filip Szczepankiewicz², Linda Knutsson^1,3, and Markus Nilsson^2
1Department of Medical Radiation Physics, Lund University, Lund, Sweden, ²Department of Diagnostic Radiology, Lund University, Lund, Sweden, ³Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, United States

Diffusion MRI can be used to probe restricted diffusion (compartment size) and water exchange (membrane permeability), but their effects on the signal are entangled. This problem can be addressed by the use of free gradient waveforms designed to disentangle the effects. However, a previously presented approach is invalid at high b-values and long diffusion encoding times. We develop a generalised restriction-exchange model that is valid for arbitrary gradient waveform and all relevant b-values and encoding times. The approach eliminates the shortcomings of the previous restriction-exchange framework.

Hossein Rafipoor¹, Saad Jbabdi¹, Ludovica Griffanti^1,2, and Michiel Cottaar^1
1Wellcome Centre for Integrative Neuroimaging (WIN), Nuffield Department of Clinical Neurosciences (NDCN), University of Oxford, Oxford, United Kingdom, ²Wellcome Centre for Integrative Neuroimaging (WIN), Department of Psychiatry, University of Oxford, Oxford, United Kingdom

We present a novel Bayesian framework to relate changes in data to changes in model parameters even in models that cannot be directly inverted. We do so by training probabilistic models that characterise how the measurements change as a result of a change in the parameters. While the approach is general, in this work we used the framework to study microstructural parameter changes that are associated with the appearance of areas of white matter hyperintensities. We found a dichotomy between periventricular and deep white matter hyperintensities, where the latter are associated with increased extracellular signal.

Tianjia Zhu^1,2, Qiyuan Tian^3,4, Susie Huang^3,4, and Hao Huang^1,5
1Department of Radiology, Children's Hospital of Philadelphia, Philadelphia, PA, United States, ²Department of Bioengineering, University of Pennsylvania, Philadelphia, PA, United States, ³Department of Radiology, Harvard Medical School, Boston, MA, United States, ⁴Massachusetts General Hospital, Boston, MA, United States, ⁵Department of Radiology, University of Pennsylvania, Philadelphia, PA, United States

Unlike diffusion in the white matter which can be well characterized by Gaussian diffusion, substantial non-Gaussian diffusion occur in the cerebral cortical mantle featured by widespread barriers of both somas and neurites. We have developed a non-Gaussian compartmental model Kurtosis-based Imaging of Neurite and Soma Architecture (KINSA) in previous studies. Here, we demonstrate the sensitivity of cortical kurtosis to diffusion time in KINSA modeling with both simulated data and in-vivo dMRI data from Connectome scanner. The mean kurtosis sensitivity to diffusion time allows for precisely delineating neurite and soma architecture such as neuronal density and soma radius.

Andreea Hertanu^1,2, Lucas Soustelle^1,2, Arnaud Le Troter^1,2, Julie Buron^1,2,3, Julie Le Priellec³, Victor N. D. Carvalho^1,2,4, Myriam Cayre³, Pascale Durbec³, Gopal Varma⁵, David C. Alsop⁵, Olivier M. Girard^1,2, and Guillaume Duhamel^1,2
1Aix Marseille Univ, CNRS, CRMBM, Marseille, France, ²APHM, Hôpital Universitaire Timone, CEMEREM, Marseille, France, ³Aix Marseille Univ, CNRS, IBDM, Marseille, France, ⁴Aix Marseille Univ, CNRS, ICR, Marseille, France, ⁵Division of MR Research, Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, United States

Inhomogeneous magnetization transfer (ihMT) signal originates from the residual dipolar interactions and is weighted by the associated dipolar relaxation time T_1D. The resulting signal can be modulated by filtering the contribution of short T_1D components to emphasize the contrast between different structures, or to enhance the specificity for myelin imaging. In this study, the dependency of ihMTR to T_1D is investigated theoretically for different T_1D-filtering strengths. Experimental WM/GM relative contrasts for the same configurations are put in perspective with theoretical contrasts resulted from single-T_1D and bi-T_1D biophysical model simulations.

Ulin Nuha Abdul Qohar¹, Erik Andreas Hanson¹, Steven Sourbron², and Antonella Zanna Munthe-Kaas^1
1Mathematics, University of Bergen, Bergen, Norway, ²University of Sheffield, Sheffield, United Kingdom

In this study, we present a simulation framework capable of generating synthetic reference perfusion MRI data suitable for evaluation and comparison of tracer kinetic models. The framework consists of a graph-based contrast agent flow model with a vascular geometry and allows for controlled simulations with realistic structural and vascular parameters. We demonstrate the potential application of the proposed framework by performing a comparison between traditional pharmacokinetic models of varying complexity, by studying the effect of ROI size.

Lucas Soustelle^1,2, Thomas Troalen³, Andreea Hertanu^1,2, Maxime Guye^1,2, Jean-Philippe Ranjeva^1,2, Guillaume Duhamel^1,2, and Olivier M. Girard^1,2
1Aix Marseille Univ, CNRS, CRMBM, Marseille, France, ²APHM, Hôpital Universitaire Timone, CEMEREM, Marseille, France, ³Siemens Healthcare SAS, Saint-Denis, France

Fast macromolecular proton fraction (MPF) mapping based on single-point (SP) quantitative MT method has shown great promises for the evaluation of myelin-related studies while allowing for acceptable scan times. The SP method requires a T₁ map, which is inherently biased by magnetization transfer effects. In this work, we investigate the effect of T₁ maps derived from different Variable Flip Angle (VFA) protocols on the computed MPF maps. It is shown that VFA-T₁ is highly variable because of MT effects, hence biasing SP-MPF maps values. The SP-MPF methodology should therefore consider MT effects in VFA-T₁ estimation, especially for cross-vendor applications.

Jonathan Doucette^1,2, Christian Kames^1,2, Christoph Birkl³, and Alexander Rauscher^1,2,4,5
1UBC MRI Research Centre, Vancouver, BC, Canada, ²Physics & Astronomy, University of British Columbia, Vancouver, BC, Canada, ³Neuroradiology, Medical University of Innsbruck, Innsbruck, Austria, ⁴Radiology, University of British Columbia, Vancouver, BC, Canada, ⁵Pediatrics, University of British Columbia, Vancouver, BC, Canada

Rapid approximate Bayesian parameter inference is investigated for $$$T_2$$$ analysis of multi spin-echo signals. We demonstrate that rapid inference using Rician noise distributions is possible using maximum likelihood estimation (MLE) if the MLE procedure is initialized with samples drawn from an approximate Bayesian posterior. This posterior is learned using a conditional variational autoencoder (CVAE) network, trained on exclusively on simulated data. Nevertheless, we show good generalization to three diverse datasets, including improved inference accuracy compared to standard nonnegative least squares-based methods which implicitly assume Gaussian noise.
 

Jelle Veraart¹, Erika P. Raven¹, Luke J. Edwards², Nikolaus Weiskopf^2,3, and Derek K. Jones^4,5
1Center for Biomedical Imaging, NYU Grossman School of Medicine, New York, NY, United States, ²Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, ³Felix Bloch Institute for Solid State Physics, Leipzig University, Leipzig, Germany, ⁴School of Psychology, Cardiff University, Cardiff, United Kingdom, ⁵Mary MacKillop Institute for Health Research, Australian Catholic University, Melbourne, Australia

The accuracy of the quantification of axon radii in vivo using diffusion MRI has been promoted in recent years by hardware developments and novel biophysical modeling insights. The MR-derived effective radii are in good quantitative agreement with histology if one accounts for the intrinsic bias of diffusion MRI to larger axons.  In this work, we show that the translation of MR axon diameter mapping to human neuroimaging is possible within acceptable scan times if strong diffusion-weighting gradients are available. Indeed, we demonstrate that the MR-derived effective axon radii is a reproducible and sensitive metric, with interesting inter- and along-tract variability.

Thomas MacLennan¹, Peter Seres¹, Julia Rickard¹, Emily Stolz¹, Christian Beaulieu¹, and Alan H. Wilman^1
1Biomedical Engineering, University of Alberta, Edmonton, AB, Canada

To perform accurate T1 and T2 mapping, $$$B_1^+$$$ maps are usually required, but not always available. For these situations, a large dataset of $$$B_1^+$$$ maps may aid in predicting $$$B_1^+$$$. In this work, Bloch-Siegert $$$B_1^+$$$ maps in the brain are characterized from a dataset of 373 healthy participants on the same 3 T. After transforming all maps to the same standard space, we show that $$$B_1^+$$$ distribution is similar across subjects with a mean CoV of 3.65% across the whole brain; slight variations were found due to brain size, shape, CSF volume, head orientation and transmit power calibration.

Ceren Tozlu¹, Keith Jamison¹, Susan Gauthier^1,2,3, and Amy Kuceyeski^1
1Department of Radiology, Weill Cornell Medicine, New York, NY, United States, ²Judith Jaffe Multiple Sclerosis Center, Weill Cornell Medicine, New York, NY, United States, ³Department of Neurology, Weill Cornell Medicine, New York, NY, United States

The complex relationship between brain’ structural connectivity (SC) and functional connectivity (FC) has not yet been fully quantified. Previous studies have shown that an increased SC-FC coupling is associated with worse cognitive performance and higher disability in people with multiple sclerosis (pwMS). However, no study to date investigated the association of regional SC-FC coupling with disability in MS. We showed that the SC-FC coupling performed a high prediction performance in classifying pwMS by impairment level. Damage to SC, particularly in the right parsorbitalis, thalamus, and parahippocampal and left superior parietal is a hallmark of disability in MS.