ISMRM 21st Annual Meeting & Exhibition 20-26 April 2013 Salt Lake City, Utah, USA

High Resolution Brain Morphometry
Tuesday 23 April 2013
Room 151 AG  13:30 - 15:30 Moderators: J. Michael Tyszka, Kei Yamada

13:30 0264.   
A Multi-Resolution Anatomical Atlas of the Human Brainstem Based on Diffusion Tensor Imaging at 11.7T
Manisha Aggarwal1, Jiangyang Zhang1, Olga Pletnikova2, Barbara Crain2, Juan Troncoso2, and Susumu Mori1
1Department of Radiology, Johns Hopkins University School of Medicine, Baltimore, MD, United States, 2Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, MD, United States

A three-dimensional anatomical atlas of the human brainstem based on ultra-high resolution (125-255 µm) DTI at 11.7T is developed. DTI contrasts in our study revealed unprecedented level of microscopic neuroanatomical details in the postmortem brainstem, comparable to histological myelin-staining. Tractography enabled reconstruction of fine fiber tracts, including interleaved fascicles of the corticospinal and transverse pontine fibers, and decussation of pyramidal tract fibers. Additionally, strong grey-white matter contrasts in ADC maps allowed precise reconstruction of grey matter nuclei. We also demonstrate mapping of the high-resolution postmortem DTI data to an in vivo whole brain atlas to construct a 3D brainstem atlas registered to the MNI stereotaxic space.

13:42 0265.   Comparison with Histology of Quantitative MR Properties of the Brain Stem Tissue in 3T and 7T
Anna I. Blazejewska1, Samuel J. Wharton1, Stefan T. Schwarz2, James Lowe3, Dorothee P. Auer2, Richard W. Bowtell1, and Penelope A. Gowland1
1SPMMRC, University of Nottingham, Nottingham, Notts, United Kingdom, 2Division of Radiological and Imaging Sciences, Nottingham University Hospitals NHS Trust, Nottingham, Notts, United Kingdom, 3Division of Pathology, Nottingham University Hospitals NHS Trust, Nottingham, Notts, United Kingdom

The substantia nigra (SN) is important in motor control and is known to be affected in Parkinson’s disease. Quantifying MR properties of this region of the brain stem and relating them to histology stains will assist in understanding tissue changes in disease and in image optimization. In this study we characterized R2* and MTR values for key regions in the brain stem at 3T and 7T. We also found correlations between those values and the appropriate post mortem stains which suggest that R2* and MTR are good indicators of iron and myelin content respectively in the brain stem.

13:54 0266.   Exploring Cortical Cytoarchitecture in High Resolution R1 Maps
José P. Marques1 and Rolf Gruetter2,3
1CIBM, University of Lausanne, Lausanne, Vaud, Switzerland, 2LIFMET - Laboratory for Functional and Metabolic Imaging, École Polytechnique Fédérale de Lausanne, Lausanne, Vaud, Switzerland, 3Department of Radiology, University of Lausanne and Geneva, Lausanne, Switzerland

In this work we explore the cytoarchitectonic information present in high-resolution (0.65mm isotropic) R1 maps at 7Tesla. We observe that R1 values decay from white matter (WM) surface to the pial surface and that the average R1 maps show a consistent qualitative agreement with the known myelin distribution throughout the brain. We further explore the spatial patterns of the R1 variation from WM to the pial surface and observe that it is possible to differentiate Brodmann areas, such as V1, from the shape of the decay.

14:06 0267.   
High-Resolution Quantitative T1 Maps of the Human Stria of Gennari at 7 Tesla
Christine Lucas Tardif1, Miriam Waehnert1, Juliane Dinse1, Andreas Schäfer1, Pierre-Louis Bazin1, and Robert Turner1
1Department of Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Saxony, Germany

The stria of Gennari, a densely myelinated tangential band in the primary visual cortex V1, has been shown in several high-resolution post-mortem and in-vivo MRI studies. We report the first estimates of T1 times in the stria of Gennari in-vivo. We scanned 2 subjects using the MP2RAGE sequence at 7 Tesla at an isotropic resolution of 0.5 mm. Using a novel volume-preserving cortical layering approach, we calculated the average cortical profile of T1 times in V1. We also sampled the cortex at the depth of the stria of Gennari and report the T1 times.

14:18 0268.   
Motion-Corrected 350 lower case Greek mum Isotropic MPRAGE at 3 T Using Volumetric Navigators (VNavs)
Matthew Dylan Tisdall1,2, Jonathan R. Polimeni1,2, and André J. W. van der Kouwe1,2
1A. A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, United States, 2Radiology, Harvard Medical School, Boston, MA, United States

350 um isotropic imaging at 3 T requires very long scan times, and subject motion is a critical impediment. We demonstrate the use of volumetric navigators (vNavs) for prospective motion and frequency correction, and the use of inner-loop GRAPPA in a MPRAGE to maintain desirable contrast as seen at 1mm. Combined, these techniques allow successful 350 um isotropic imaging of a healthy volunteer without additional restraint.

14:30 0269.   A Simple Method to Denoise MP2RAGE
Kieran O'Brien1, Gunnar Krueger2,3, François Lazeyras1, Rolf Gruetter1,4, and Alexis Roche2,5
1CIBM, University of Geneva, Geneva, Switzerland, 2Advanced Clinical Imaging Technology, Siemens Healthcare IM S AW, Lausanne, Switzerland, 3CIBM, Ecole Polytechnique Fédérale de Lausanne & University of Lausanne, Lausanne, Switzerland, 4CIBM, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland,5CIBM, CHUV, Lausanne, Switzerland

MP2RAGE provides a self-bias corrected image by taking the image ratio of two image volumes; however due to the inherent numerical instability of a ratio calculation, this causes the signal to diverge when low and the noise in the ratio image to be amplified. To remove the numerical instability we modified the image ratio calculation with a tunable parameter, gamma, that can be optimized for individual data sets to suppress noise without affecting voxels with signal. When applied to MP2RAGE scans, the ratio results in customary T1w images.

14:42 0270.   How Much Resolution Is Needed for in-vivo Analysis of Cortical Myeloarchitecture?
Miriam Waehnert1, Juliane Dinse1,2, Christine Lucas Tardif1, Andreas Schäfer1, Stefan Geyer1, Pierre-Louis Bazin1, and Robert Turner1
1Department of Neurophysics, Max-Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, 2Faculty of Computer Science, OvGU Magdeburg, Magdeburg, Germany

Intra-cortical T1 contrast reflects myeloarchitecture in the living human brain. Profiles can be constructed on MRI of quantitative T1 maps in 3D and used to detect cortical areal boundaries. Here we explore the effect of spatial resolution on the distinguishability of cortical architecture. We compare cortical profiles from the neighbouring Brodmann areas 1, 3b and 4 from T1 maps with isotropic resolutions of 0.5 mm and 0.7 mm. The standard deviations of the average profiles are mostly smaller at higher resolution. Moreover, the shapes of cortical profiles are similar at 0.7 mm, whereas they are Brodmann-area specific at 0.5 mm.

14:54 0271.   The Fixation Protocol Alters Brain Morphology in ex-vivo MRI Mouse Phenotyping
A. Elizabeth de Guzman1,2, Michael D. Wong1,2, Jacqueline A. Gleave1, and Brian J. Nieman1,2
1Mouse Imaging Centre, Hospital for Sick Children, Toronto, Ontario, Canada, 2Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada

Despite the common use of chemically fixed tissue for MRI morphometric analysis of the rodent brain, limited details are available describing how the fixation protocol may alter brain morphology. In this study, fixed mouse brain samples were imaged repeatedly during fixation in order to determine the effect that changes in fixation time has on structure volume. Each structure in the brain changed in volume at different rates with fixation time, where structures near the ventricular system decreased in size while the ventricles expanded. Caution should be taken to maintain consistency with fixation protocols, as within study changes may masquerade as phenotypic differences.

15:06 0272.   MRI of Cell Layers in Mouse Brain in Vivo Using Intra- And Extra-Cellular Contrast Agents
Takashi Watanabe1, Jens Frahm1, and Thomas Michaelis1
1Biomedizinische NMR Forschungs GmbH am Max-Planck-Institut für biophysikalische Chemie, Goettingen, Germany

This work demonstrates a use of manganese or Gd-DTPA for MRI of the olfactory bulb, hippocampus, and cerebellum of anesthetized mice. Systemic manganese administration improves the contrast between tightly packed cellular layers and surrounding layers not only in T1-weighted but also in T2-weighted MRI. Intracranial Gd-DTPA administration reverses the contrast in T1-weighted MRI and generates a contrast similar to that in T2-weighted MRI. With manganese or Gd-DTPA, 12-min T1-weighted MRI provides a higher contrast-to-noise ratio than 161-min T2-weighted MRI at 30×30×300 µm3 resolution. The contrast-enhanced MRI at 25×25×250 µm3 shows anatomical contrast in unprecedented ways.

15:18 0273.   Magic Angle Enhanced MR Microscopy of Fibrous Structures in the Eye
Kevin C. Chan1,2, Zion Tse3, Ning-Jiun Jan4, Joel S. Schuman2, Seong-Gi Kim1,5, and Ian A. Sigal2,4
1Neuroimaging Laboratory, University of Pittsburgh, Pittsburgh, Pennsylvania, United States, 2Departments of Ophthalmology and Bioengineering, University of Pittsburgh, Pittsburgh, Pennsylvania, United States, 3College of Engineering, University of Georgia, Athens, Georgia, United States, 4Laboratory of Ocular Biomechanics, University of Pittsburgh, Pittsburgh, Pennsylvania, United States, 5Department of Radiology, University of Pittsburgh, Pittsburgh, Pennsylvania, United States

This study explored the use of high-field magic-angle enhanced MR microscopy to evaluate the layer-specific tissue properties of the fibrous microstructures in the eye using the 9.4T scanner. Distinct fibrous microstructures and differential T2*-weighted signal intensity profiles were observed layer-specifically in the anterior and posterior sclera, cornea, lens and optic nerve head at different orientations to Bo. When orientating the tissue samples from 0o to 90o relative to Bo, maximum signal intensity was found for all sclera, cornea and tendon samples at the magic angle (55o to Bo) by 82%, 24% and 220% respectively. The results of this study may open up new areas on non-invasive assessments of biomechanical and biochemical properties of collagen fiber distribution and deformation and remodeling in the eye, and may potentiate future studies on longitudinal monitoring of functional microstructures in diseases involving the corneoscleral shell and optic nerve fibers such as glaucoma, myopia and aging.