Room 251 BCEF

10:30 - 12:30

Moderators:

Jonathan D. Clayden, Jennifer A. McNab

Shani Ben Amitay¹ and Yaniv Assaf^1
1Department of Neurobiology, Tel Aviv University, Tel Aviv, Israel

Parcellation of the cortex is traditionally achieved by visualization of the cortical cellular architecture. In this work we aim to segment the cortex into neuro-anatomical regions based on variations in restricted diffusion properties along the cortex. We use hierarchical cluster analysis (HCA) to segment the cortex based on DTI and CHARMED maps. HCA and surface based analysis of the restricted diffusion component (Fr) indicates that homologous regions share similar mean Fr values and that mean Fr differs among cortical regions. We find that CHARMED seems to be superior over DTI parameters for cortical characterization.

Stamatios N. Sotiropoulos¹, Charlie Chen², Krikor Dikranian², Saad Jbabdi¹, Timothy E.J. Behrens¹, David C. Van Essen², and Matthew F. Glasser^2
1FMRIB Centre, University of Oxford, Oxford, Oxfodshire, United Kingdom, ²Department of Anatomy & Neurobiology, Washington University, St Louis, MO, United States

Despite the unique information that can be extracted from diffusion-weighted (DW) MRI of the living brain, validation is limited by the inherent difficulties in obtaining ground truths. Therefore, many studies have been recently conducted for developing challenging validation frameworks. Such a validation setup is presented in this abstract. We utilise histological preparation and myelin staining of slices of a macaque brain to map the neuronal fibre organisation. We then use image processing approaches to extract ground truths on certain anatomical features of the underlying fibres. Finally, we compare these features with predictions from model-based analysis of macaque DW-MRI scans.

Trong-Kha Truong¹, Arnaud Guidon¹, and Allen W. Song^1
1Brain Imaging and Analysis Center, Duke University, Durham, NC, United States

This study investigates the cortical depth dependence of diffusion anisotropy in the human cortical gray matter in vivo at 3T. A novel multi-shot constant-density spiral diffusion tensor imaging technique with inherent correction of motion-induced phase errors was used to achieve a high spatial resolution, signal-to-noise ratio, and spatial fidelity. The results show that cortical gray matter has a different principal eigenvector orientation as compared to that of adjacent white matter regions as well as a clear cortical depth dependence of the fractional anisotropy, with consistently higher values in the middle cortical layers than in the superficial and deep cortical layers.

Michiel Kleinnijenhuis^1,2, Hui Zhang³, Dirk Wiedermann⁴, Benno Küsters⁵, Anne-Marie van Cappellen van Walsum^1,6, and David G. Norris^2,6
1Department of Anatomy, Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, ²Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, Nijmegen, Netherlands, ³Department of Computer Science & Centre for Medical Image Computing, University College London, London, United Kingdom, ⁴Max Planck Institute for Neurological Research, Cologne, Germany, ⁵Department of Pathology, Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, ⁶MIRA Institute for Biomedical Technology and Technical Medicine, Enschede, Netherlands

We demonstrate that, using the multi-compartment tissue model NODDI, a fine-grained description of the laminae of the human neocortex can be achieved outperforming diffusion tensor metrics.

Manisha Aggarwal¹, Jiangyang Zhang¹, and Susumu Mori^1
1Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, United States

We investigate the effect of gradient frequency on diffusion tensor contrasts derived using oscillating-gradient diffusion MRI in the postmortem fixed human hippocampus. Using DTI data acquired at 11.7T with pulsed-gradient and oscillating-gradient diffusion sensitizing gradient waveforms (with modulation frequencies of 50, 100, and 150 Hz), distinct frequency-dependent contrasts were observed in specific layers of the human hippocampus. ADC, parallel and perpendicular diffusivity measurements showed progressive region-specific increases with increasing gradient frequency. The results demonstrate that selective sampling of the diffusion spectrum with oscillating-gradient diffusion MRI can generate contrasts that are uniquely sensitive to the cytoarchitectural heterogeneity of different layers in the hippocampus.

Shir Hofstetter^1,2, Naama Friedmann^1,3, and Yaniv Assaf^1,2
1Sagol School of Neuroscience, Tel Aviv University, Tel Aviv, Israel, ²Neurobiology, Tel Aviv University, Tel Aviv, Israel, ³School of Education, Tel Aviv University, Tel Aviv, Israel

Recent DTI study found changes in diffusion indices after 2 hour training. Whether short learning experiences can induce microstructural alterations in the neocortex is unclear and requires high scanning resolution. Here we used diffusion MRI to study cortical plasticity by a short (less than 1 hour) language task (introducing new lexical items to the lexicons). Changes in diffusion were found in specific cortical region involved in language and reading. These results provide first indication of micro-structural changes in the cortex induced by a short task with high cognitive demands, and the ability of diffusion MRI to reveal such rapid modifications.

Gernot Reishofer¹, Christian Langkammer², David Porter³, Karl Koschutnig⁴, Margit Jehna¹, Robert Merwa⁵, and Franz Ebner^1
1Radiology, Medical University of Graz, Graz, Austria, ²Neurology, Medical University of Graz, Graz, Austria, ³MR R&D, Siemens AG, Healthcare Sector, Erlangen, Germany, ⁴Institute of Psychology, Universtiy of Graz, Graz, Austria, ⁵Medical Engineering, Upper Austria University of Applied Sciences, Linz, Austria

Readout-segmented echo planar imaging (RS-EPI) with 2D navigator-based reacquisition is a new promising technique that enables the sampling of high resolution DWI with reduced susceptibility artifacts. Giving the fact that long scan times are required for high SNR due to the measurement of two or more averages, we set out to regularize the diffusion tensor from RS-EPI DWI based on one measurement. The proposed total variation based regularization algorithm is user independent since the regularization parameter is evaluated automatically and accounts for spatial varying SNR. Tractography from regularized data show a significant improvement compared to unregularized data.

David A. Raffelt¹, Robert E. Smith¹, Jacques-Donald Tournier^1,2, Gerard R. Ridgway³, Victor L. Villemagne⁴, Christopher C. Rowe⁴, Olivier Salvado⁵, and Alan Connelly^1,2
1Advanced MRI Development, Florey Institute of Neuroscience and Mental Health, Melbourne, VIC, Australia, ²Department of Medicine, University of Melbourne, Melbourne, VIC, Australia, ³Wellcome Trust Centre for Neuroimaging, UCL Institute of Neurology, London, England, United Kingdom,⁴Department of Nuclear Medicine, Austin Health, Melbourne, VIC, Australia, ⁵The Australian E-Health Research Centre, CSIRO, Brisbane, QLD, Australia

Recent work has focused on the development of more interpretable quantitative diffusion MRI measures that can be associated with a specific fibre bundle within a voxel containing crossing fibres. Unfortunately, traditional methods for voxel-based analysis of 3D images cannot be applied to analyse these measures, since the definition of the local neighbourhood for smoothing and cluster construction is ambiguous when adjacent voxels may contain different fibre populations. We propose a novel statistical framework for whole-brain voxel-based analysis of bundle-specific diffusion MRI measures, which uses group-average fibre tractography to define the local neighbourhood for clustering and smoothing purposes.

Sjoerd B. Vos¹, Murat Aksoy², Zhaoying Han², Samantha J. Holdsworth², Christoph Seeger², Maclaren Julian², Alexander Brost², Alexander Leemans¹, and Roland Bammer^2
1Image Sciences Institute, University Medical Center Utrecht, Utrecht, Netherlands, ²Center for Quantitative Neuroimaging, Department of Radiology, Stanford University, Stanford, California, United States

By combining advanced 3D DW EPI acquisitions with high-angular resolution diffusion imaging we have created a unique DWI dataset, of 1mm true isotropic resolution with 60 diffusion weighted directions. This dataset was reconstructed to create subsets of low angular (15 and 30 directions) and lower spatial (2mm isotropic) resolution, to compare the influence of both of these factors on fiber tractography, both in a qualitative, visual, manner as well as quantitatively.

Silvia De Santis¹, Yaniv Assaf², Sonya Bells¹, Sean C. L. Deoni³, and Derek K. Jones^1
1CUBRIC Cardiff University, Cardiff, United Kingdom, ²Department of Neurobiology, Tel Aviv University, Tel Aviv, Israel, ³School of Engineering, Brown University, Providence, RI, United States

We have developed the first microstructural atlas that includes both DT-MRI and advanced microstructural indices. The atlas includes population means and standard deviations of multi-variate data for every voxel in MNI space, and for all major association, projection and commissural pathways which can be described and differentiated on the basis of their microstructural properties, and is meant to serve as a reference for all the experimenters interested in microstructural imaging.