Kurt Schilling¹, Yurui Gao¹, Vaibhav Janve¹, Iwona Stepniewska², Prasanna Parvathaneni³, Hua Li¹, Bennett A Landman⁴, and Adam W Anderson^1
1VUIIS, Vanderbilt University, Nashville, TN, United States, ²Psychology, Vanderbilt University, Nasvhille, United States, ³Electrical Engineering, Vanderbilt University, TN, United States, ⁴Electrical Engineering, Vanderbilt University, Nashville, TN, United States

Validation of fiber orientation and anisotropy estimates from diffusion MRI (dMRI) studies has so far been limited to 2D analysis of histological sections. Here, we use structure tensor analysis to extract the 3D histological fiber orientation distribution (FOD) from confocal z-stacks, which then serves as a gold standard for comparisons with various dMRI acquisition protocols and signal models. Our results show good agreement between histological and dMRI estimates of fiber orientation in voxels containing nearly parallel fibers. Further, we find that Q-ball and DSI analysis are also able to resolve crossing fibers with similar accuracy.

Matthan W.A. Caan¹, Ezequiel Farrher², James Cole³, Dirk H.J. Poot^4,5, Farida Grinberg^2,6, and N. Jon Shah^2,6
1Department of Radiology, Academic Medical Center, Amsterdam, Netherlands, ²Institute of Neuroscience and Medicine-4, Forschungszentrum Juelich, Juelich, Germany, ³Computational, Cognitive, and Clinical Neuroimaging Laboratory, Division of Brain Sciences, Imperial College London, London, United Kingdom, ⁴Quantitative Imaging Group, Department of Imaging Physics, Delft University of Technology, Delft, Netherlands, ⁵Biomedical Imaging Group Rotterdam, Erasmus MC, Rotterdam, Netherlands, ⁶Department of Neurology, Faculty of Medicine, JARA, RWTH Aachen University, Aachen, Germany

We studied diffusion parameters within a multisection diffusion phantom with parallel, diverging and crossing fibers, of which HARDI data were acquired on two 3 Tesla scanners. The single tensor model, a constrained dual tensor (DT) model and diffusion kurtosis imaging (DKI) were employed. Diffusion parameter profiles were computed in fibers tracked in different parts of the phantom. The DT model and DKI were more sensitive to minor microstructural changes. Comparing data between scanners shows an offset smaller than one standard deviation in all parameters. This approach gives insights into which parameters are most appropriate for multi-site human studies.

Damien Jacobs¹, Benoit Scherrer², Aleksandar Jankovski³, Anne des Rieux⁴, Maxime Taquet¹, Bernard Gallez⁴, Simon K. Warfield², and Benoit Macq^1
1ICTEAM, Universite catholique de Louvain, Louvain-La-Neuve, Belgium, ²Computational Radiology Laboratory, Boston Childrens Hospital, Massachusetts, United States, ³Hopital universitaire Mont-Godinne, Universite catholique de Louvain, Godinne, Belgium, ⁴LDRI, Universite catholique de Louvain, Brussels, Belgium

The purpose is to investigate, for diffusion compartment models (NODDI and DIAMOND), the variations of diffusion parameters after Wallerian degeneration in the rat spinal cord. The values of diffusion parameters for NODDI and DIAMOND provide different informations and both models are investigated in the case of the Wallerian degeneration process. The alterations of each diffusion parameters are highlighted after an unilateral rhizotomy and the results are compared with the histological observations (myelin, neurofilaments, microglia, astrocytes).

Francesco Grussu¹, Torben Schneider¹, Richard L. Yates², Mohamed Tachrount³, Hui Zhang⁴, Daniel C. Alexander⁴, Gabriele C. DeLuca², and Claudia A. M. Wheeler-Kingshott^1
1NMR Research Unit, Department of Neuroinflammation, Queen Square MS Centre, UCL Institute of Neurology, London, England, United Kingdom,²Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, England, United Kingdom, ³Department of Brain Repair and Rehabilitation, UCL Institute of Neurology, London, England, United Kingdom, ⁴Department of Computer Science and Centre for Medical Image Computing, University College London, London, England, United Kingdom

Neurite orientation dispersion is an important morphological feature at the MRI voxel scale even in coherent areas such as the corpus callosum. In this work, we investigate its importance in another organised area: the human spinal cord. We estimate orientation dispersion in an ex vivo specimen with neurite orientation dispersion and density imaging (NODDI) at 9.4 T, and compare results with structure tensor features of silver stained sections from the same sample. We conclude that orientation dispersion is a key microstructural characteristic also of the spinal cord, and NODDI is a reliable technique for its quantification.

Henrik Lundell¹, Tim B. Dyrby¹, Penny L. Hubbard Cristinacce^2,3, Feng-Lei Zhou^2,4, Geoffrey J.M. Parker^2,3, and Sune N. Jespersen^5,6
1Centre for Functional and Diagnostic Imaging and Research, Copenhagen University Hospital, Hvidovre, Denmark, ²Centre for Imaging Sciences, The University of Manchester, United Kingdom, ³Biomedical Imaging Institute, The University of Manchester, United Kingdom, ⁴The School of Materials, The University of Manchester, United Kingdom, ⁵CFIN/MINDLab, Aarhus University, Denmark, ⁶Department of Physics and Astronomy, Aarhus University, Denmark

Double diffusion encoding pulse sequences (also known as double PFG) have been shown to detect microscopic diffusion anisotropy in macroscopically isotropic systems. Here we present and validate three sets of gradient direction schemes to measure microscopic diffusion anisotropy (e.g. µFA) in the presence of macroscopic anisotropy. In a phantom of co-electro-spun fibers, our results demonstrate similar performance of the three schemes, and show how µFA and FA together inform about microscopic anisotropy and fiber dispersion in a model independent fashion.

Damien J McHugh^1,2, Fenglei Zhou^1,3, Penny L Hubbard Cristinacce^1,2, Josephine H Naish^1,2, and Geoff J M Parker^1,2
1Centre for Imaging Sciences, The University of Manchester, Manchester, United Kingdom, ²Biomedical Imaging Institute, The University of Manchester, Manchester, United Kingdom, ³Materials Science Centre, The University of Manchester, Manchester, United Kingdom

In this work we introduce a novel phantom with microstructural characteristics mimicking tumour cellular structure, consisting of a collection of roughly spherical polymer structures. A two-compartment biophysical model was applied to diffusion data acquired over a range of gradient strengths and diffusion times, allowing the size and volume fraction of the ‘cells’ to be estimated. The radius was slightly underestimated compared with that determined from scanning electron microscope measurements, the fitted volume fraction was plausible, and parameters were found to be estimated with reasonably good precision. Such phantoms may be useful for testing microstructural models of the diffusion signal, and for scanner and protocol calibration.

Johannes Lindemeyer¹, Ezequiel Farrher¹, Farida Grinberg^1,2, Ana-Maria Oros-Peusquens¹, and N. Jon Shah^1,2
1Institute of Neuroscience and Medicine 4, INM-4, Medical Imaging Physics, Forschungszentrum Jülich GmbH, Jülich, Germany, ²Faculty of Medicine, Department of Neurology, RWTH Aachen University, JARA, Aachen, Germany

In this work we present an approach to reduce the effect of microscopic, susceptibility-induced field gradients on diffusion metrics in artificial anisotropic fibre phantoms. The approach relies on the use of MgCl2 to match the magnetic susceptibility of the liquid to that of the fibres.

Alexandru V. Avram¹, Michal E. Komlosh^1,2, Alan S. Barnett^1,2, Elizabeth Hutchinson^1,2, Dan Benjamini^1,3, and Peter J. Basser^1
1Section on Tissue Biophysics and Biomimetics, NICHD, National Institutes of Health, Bethesda, MD, United States, ²The Henry Jackson Foundation, Bethesda, MD, United States, ³Department of Biomedical Engineering, Tel-Aviv University, Tel-Aviv, Israel

Experimental design and environmental factors can bias the quantitation of anisotropic diffusion using DTI-derived metrics, such as FA, compromising the value of longitudinal data and multicenter clinical studies. We propose the use of a novel anisotropic diffusion phantom in conjunction with a general method for modeling the diffusion signal produced by that phantom using the details of the DWI pulse sequence and the multiple correlation function (MCF) framework. The well-defined and known microstructure of the phantom generates a wide range of DTI parameters that can be used to calibrate various DTI pulse sequences and optimize clinical DTI protocols.

Xiaoke Wang¹, Scott B. Reeder^2,3, and Diego Hernando^2
1Biomedical Engineering, University of Wisconsin, Madison, Wisconsin, United States, ²Radiology, University of Wisconsin, Madison, Wisconsin, United States, ³Medical Physics, University of Wisconsin, Madison, Wisconsin, United States

Phantoms exhibiting desired diffusion behavior could provide a controlled means for validation of quantitative diffusion MRI techniques. Ideally, such phantoms should possess a single-peak MR spectrum and ADC values over the entire physiological range at a controlled temperature. Recently proposed phantoms such as PVP and sucrose phantoms may not meet all of these requirements. The purpose of this study is to investigate the spectral diffusion behavior of sucrose and PVP phantoms, and propose acetone D2O mixture as an alternative. Sucrose phantom showed multiple spectral peaks. PVP phantom showed narrow ranged ADC. Acetone-D2O phantom meets the requirements described above.

Juan Parra-Robles¹, Bart Veeckmans², Madhwesha Rao¹, James C Hogg³, and Jim M Wild^1
1University of Sheffield, Sheffield, South Yorkshire, United Kingdom, ²Materialise, Leuven, Belgium, ³University of British Columbia, Vancouver, British Columbia, Canada