HARDI & Higher Order Descriptions of Diffusion
Tuesday 21 April 2009
Room 316BC 16:00-18:00


Daniel Alexander and Mariana Lazar

16:00  357. Orientationally Invariant Axon-Size and Density Weighted MRI
    Daniel C. Alexander1, Penny L. Hubbard2, Matt G. Hall1, Elizabeth A. Moore3, Maurice Ptito4, Geoff J. M. Parker2, Tim B. Dyrby5
Centre for Medical Image Computing, Dept. Computer Science, UCL (University College London), London, UK; 2Imaging Science and Biomedical Engineering, School of Cancer and Imaging Sciences, University of Manchester, Manchester, UK; 3Best, Philips Healthcare, Eindhoven, Netherlands; 4School of Optometry, University of Montreal, Montreal, Canada; 5Danish Research Centre for Magnetic Resonance, Copenhagen University Hospital, Copenhagen, Denmark
    We present a diffusion MRI technique for mapping correlates of the axon size and density in white matter over the whole brain. Unlike previous techniques, the method does not require prior knowledge of the fibre direction. Results from a perfusion fixated monkey brain agree with histological data in the literature. Further results from in vivo human brain data, acquired in less than one hour on a standard system, also show the expected trends.
16:12 358. How Many Diffusion Gradient Directions Are Required for HARDI?

J-Donald Tournier1,2, Fernando Calamante1,2, Alan Connelly1,2
Brain Research Institute, Florey Neuroscience Institutes (Austin), Melbourne, Victoria, Australia; 2Department of Medicine, University of Melbourne, Melbourne, Victoria, Australia

    While the number of DW directions required for DTI has been extensively studied, the equivalent studies have not been performed for HARDI. However, due to the large number of directions inherent in HARDI acquisitions, this information is crucial to keep scan times practical. In this study, we use sampling theory to determine the minimum number of DW directions required for HARDI experiments, independent of the particular algorithm used to estimate fibre orientations. Results obtained at b = 3000s/mm2 indicate that 45 directions are required to adequately characterize the angular frequency information contained in the DW signal.
16:24 359.

Diffusion Kurtosis Imaging (DKI) Reveals an Early Phenotype (P30) in a Transgenic Rat Model for Huntington’s Disease

    Ines Blockx1, Marleen Verhoye1,2, Geert De Groof1, Johan Van Audekerke1, Kerstin Raber3, Dirk Poot2, Jan Sijbers2, Stephan von Horsten3, Annemie Van der Linden1
Bio-Imaging Lab, University of Antwerp, Antwerp, Belgium; 2Vision Lab, University of Antwerp, Antwerp, Belgium; 3Experimental Therapy, Friedrich-Alexander University, Erlangen, Germany
    The transgenic rat model for Huntington’s Disease (HD) resembles the late-onset form, exhibiting a behavioral phenotype with emotional disturbance, motor deficits, and cognitive decline. We used Diffusion Kurtosis Imaging (DKI) to study very young (P15-P30) HD rats and controls. In white matter of P30 HD pups we observed a significantly decreased fractional anisotropy and increased axial kurtosis as compared to controls. In grey matter of P30 HD pups axial kurtosis was significantly increased as well as Mean Kurtosis. The DKI changes we detected in HD pups, suggests that neurodegenerative processes of HD also involves neurodevelopment defects already detectable at P30.


16:36 360. In Vivo Imaging of Kurtosis Tensor Eigenvalues in the Brain at 3 T
    Eric Edward Sigmund1, Mariana Lazar1, Jens H. Jensen1, Joseph A. Helpern1
Radiology, New York University, New York, NY , USA
    Diffusion kurtosis imaging (DKI) quantifies the well-known non-gaussianity of apparent diffusion in biological tissue, and its directionally averaged mean kurtosis (MK) has potential clinical utility in the brain. However, the complete rank-4 kurtosis tensor contains more information that may be reduced, as in diffusion tensor diagonalization, to characteristic properties. The present study applies the recently described spectral decomposition to derive the eigenvalues/eigentensors of the kurtosis tensor, and resolve them spatially in a full brain 3 T DW-MRI scan. Maps of kurtosis eigenvalues and composite “eigensurfaces” derived from eigentensor projections present new contrast, potentially involving fiber crossing and barrier-density anisotropy.
16:48 361. Detection of Brain Maturation - DTI with Different B-Values Versus Diffusion Kurtosis Imaging
    Matthew Man Hin Cheung1,2, Ed X. Wu1,2
Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong SAR, China; 2Laboratory of Biomedical Imaging and Signal Processing , The University of Hong Kong, Hong Kong SAR, China
    This study aims to investigate the effects of different b-values in detecting microstructural changes during well-controlled rodent brain maturation. The results are also compared to the diffusion kurtosis indices obtained by diffusion kurtosis imaging (DKI) that characterizes the restricted diffusion by fitting multiple b-value DW measurements to a quadratic-exponential model. The results indicate that the b-value for optimal DTI detection of microstructural changes depends on the specific physiological or pathological processes targeted. High-order diffusion imaging, such as DKI, is therefore essential for more robust MR diffusion characterization of neural tissues.
17:00 362. Super-Resolving Patches in Diffusion MRI Using Canonical Fibre Configurations
    Simon Jeremy Damion Prince1, Daniel C. Alexander1
Computer Science, University College London, London, UK
    We aim to improve the spatial resolution of fibre-orientation estimates in diffusion MRI by learning a prior over 3x3x3 voxel patches that assumes self-similarity across scale. In training we align and cluster patches of fibre orientations from a larger scale using a mixture of Watsons model. The resulting canonical fibre configurations describe homogenous regions, bending, fanning etc. To super-resolve, we find the canonical patch and mean orientation, diffusivity and volume fraction that best describe the voxel measurements. We compare our results to nearest-neighbour interpolation and demonstrate that it is possible to successfully determine sub-voxel structure.
17:12 363. Model-Based Residual Bootstrap of Constrained Spherical Deconvolution for Probabilistic Segmentation and Tractography
    Hamied Ahmad Haroon1, David M. Morris1, Karl V. Embleton1,2, Geoff J. Parker1
Imaging Science and Biomedical Engineering, School of Cancer and Imaging Sciences, The University of Manchester, Manchester, England, UK; 2School of Psychological Sciences, The University of Manchester, Manchester, England, UK
    Here we describe the application of model-based residual bootstrapping to the analysis of HARDI data using constrained spherical deconvolution. We demonstrate that the method is able to provide estimates of the probability of finding different fiber configurations within the brain. These distributions of fiber orientations may then be used directly as PDFs across each configuration for probabilistic tractography. This method provides a means by which the microstructural complexity of tissue, as reflected in the HARDI diffusion signal, may be characterised, naturally accounting for the underlying tissue microscopic complexity, macroscopic partial volume, and data noise levels.
17:24 364. Diffusion Propagator Imaging: A Novel Technique for Reconstructing the Diffusion Propagator from Multiple Shell Acquisitions
    Maxime Descoteaux1, Jean-François Mangin1, Cyril Poupon1
NeuroSpin, IFR 49, CEA Saclay, Gif-sur-Yvette, France
    We present a simple and linear analytical diffusion propagator solution using spherical Laplace's equation. The reconstruction is possible from only two b-value HARDI acquisitions and less than 100 diffusion measurements.
17:36 365. Practical Crossing Fiber Imaging with Combined DTI Datasets and Generalized Reconstruction Algorithm
    Fang-Cheng Yeh1, Van Jay Wedeen2, Wen-Yih Isaac Tseng1,3
Center for Optoelectronic Biomedicine, National Taiwan University College of Medicine, Taipei, Taiwan; 2MGH Martinos Center for Biomedical Imaging, Harvard Medical School, Charlestown, MA, USA; 3Department of Medical Imaging, National Taiwan University Hospital, Taipei, Taiwan
    We present a clinically feasible sampling scheme which combines two DTI datasets to achieve higher angular resolution. The performance and accuracy of the proposed scheme was examined in comparison with q-ball imaging method.
17:48 366. Restriction Spectrum Imaging (RSI): A New Method for Resolving Complex Tissue Microstructures in Diffusion MRI
    Nathan S. White1, Trygve B. Leergaard2, Alex de Crespigny3,4, Anders M. Dale5
Cognitive Science, University of California, San Diego, La Jolla, CA, USA; 2Centre for Molecular Biology and Neuroscience, University of Oslo, Norway; 3Radiology and Neurosciences, Massachusetts General Hospital, Harvard University; 4Department of Clinical Neurology, Oxford University, UK; 5Neurosciences and Radiology, University of California, San Diego, La Jolla, CA, USA
    We present a new model-based analysis method for diffusion MRI called Restriction Spectrum Imaging (RSI). Traditional model-based deconvolution methods assume all fibers in a given voxel are composed of the same biological components and thus have identical water restriction properties. In RSI, the tissue is modeled using a spectrum of both oriented and non-oriented diffusion components with different water restriction scales. Both the volume fraction and 3D orientation of each individual component can be derived in each voxel using linear estimation methods. RSI may provide a more complete characterization of tissue with complex neuromorphologies.