Diffusion Biophysics & Microstructural Imaging

Monday 12 May 2014

Noam Shemesh¹, Jens T Rosenberg^2,3, Jean-Nicolas Dumez¹, Samuel Colles Grant^2,3, and Lucio Frydman^1,2
1Chemical Physics, Weizmann Institute of Science, Rehovot, Israel, ²National High Magnetic Field Laboratory, The Florida State University, Tallahassee, FL, United States, ³Chemical & Biomedical Engineering, The Florida State University, Tallahassee, FL, United States

Microstructural characterizations of the Central Nervous System are limited by water’s non-specificity. Whereas diffusion Magnetic Resonance Spectroscopy (MRS) methods utilize a spectral dimension to impart specificity, they still rely on detection of Apparent Diffusion Coefficients (ADCs), whose relationship with underlying microstructural determinants are tenuous. Here, we combine a recent spectrally-specific Longitudinal Relaxation Enhancement (LRE) approach with the double-Pulsed-Field-Gradient (dPFG) approach that provides unambiguous and clear indications for restricted diffusion. The ensuing localized dPFG MRS provides unique signatures for metabolic confinements, which are shown to vary in stroked rats. These unique measurements provide a glimpse into the nature of ADC variations in stroke.

Filip Szczepankiewicz¹, Samo Lasic², Jimmy Lätt³, Danielle van Westen⁴, Carl-Fredrik Westin⁵, Freddy Ståhlberg^1,4, Daniel Topgaard⁶, and Markus Nilsson^7
1Department of Medical Radiation Physics, Lund University, Lund, Sweden, ²CR Development, AB, Lund, Sweden, ³Center for Medical Imaging and Physiology, Skane University Hospital, Lund, Sweden, ⁴Department of Diagnostic Radiology, Skane University Hospital, Lund, Sweden, ⁵Laboratory of Mathematics in Imaging, Harvard Medical School, Boston, MA, United States, ⁶Center for Chemistry and Chemical Engineering, Lund University, Lund, Sweden, ⁷Lund University Bioimaging Center, Lund University, Lund, Sweden

In this work we present the first in vivo experiments employing magic angle spinning of the q-vector (qMAS) to map the microscopic anisotropy of the brain. This technique allows for the parameterization of anisotropy that is unaffected by the orientation dispersion. This means that the anisotropy is probed on a sub-voxel scale, and can potentially be useful in complex white matter geometries and gray matter, where conventional metrics such as FA are confounded by the tissue micro architecture.

Tanguy Duval¹, Jennifer A. McNab², Kawin Setsompop³, Thomas Witzel³, Torben Schneider⁴, Susie Yi Huang², Boris Keil³, Eric Klawiter³, Lawrence L. Wald³, and Julien Cohen-Adad^1
1Institute of Biomedical Engineering, Polytechnique Montreal, Montreal, Quebec, Canada, ²Department of Radiology, Stanford University, Stanford, California, United States, ³A.A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School, Charlestown, Massachusetts, United States, ⁴NMR Research Unit, Department of Neuroinflammation, Queen Square MS Centre, UCL Institute of Neurology, London, London, United Kingdom

Composite hindered and restricted model of diffusion was shown to retrieve white matter micro-structural information, such as axon diameter. Using a dedicated human gradient system that can achieve 300 mT/m, we produced the first in vivo mapping of axon diameter in the human spinal cord. State-of-the-art methods were deployed to overcome the numerous artifacts associated with spinal cord imaging. Maps of axon diameter were generated for each patient, and diffeomorphic registration on a white-matter template yielded an average atlas of axon diameter. Qualitative comparison with histological data on a rat suggests consistent trends of axon diameter across specific spinal pathways.

Daniel Barazany^1,2, Derek K Jones², and Yaniv Assaf^1
1Department of Neurobiology, Tel Aviv University, Tel Aviv, Israel, ²CUBRIC School of Psychology, Cardiff University, Cardiff, United Kingdom

AxCaliber is a diffusion MRI method that models the axon diameter distribution. here we introduce a noval acquisition that Combines T1 and AxCaliber to provide further insight into white matter micro-structure as it links between axons at different size and their T1. This kind of combined acquisition and analysis allows to estimate the effect of myelin content on T1 and its relation to axon diameter. The results of this work demonstrate that MRI allows to probe white matter tissue micro-structure providing invaluable high details that so far could be extracted only by invasive techniques.

Alice Lebois¹, Chun Hung Yeh², Denis Le Bihan³, Ching-Po Lin², and Cyril Poupon^3
1NeuroSpin/CEA, Gif-Sur-Yvette, France, ²National Yang-Ming University, Taiwan, ³NeuroSpin/CEA, France

We here propose to study the reason for the systematic overestimation of the smaller radius in available axon diameter mapping techniques when a simple cylinder model is used for the axon. Recent studies have introduced an alternative model, assuming that water molecules close to the axon membranes have a slow diffusivity while those far from the membranes are characterized by a fast diffusivity, yielding the two-pool cylinder model. This study shows how attenuations from this model when varying the thickness of the layer close to axonal membranes can be similar to attenuations from the simple cylinder model with higher radius for small radii.

Nikola Stikov¹, Jennifer S.W. Campbell¹, Mariette Lavallée¹, Thomas Stroh¹, Stephen Frey¹, Jennifer Novek¹, Stephen Nuara¹, Ming-Kai Ho¹, Barry Bedell¹, and G. Bruce Pike^1,2
1Montreal Neurological Institute, McGill University, Montreal, QC, Canada, ²Hotchkiss Brain Institute, University of Calgary, Calgary, AB, Canada

The myelin g-ratio, defined as the ratio between the inner and the outer diameter of the myelin sheath, is a fundamental property of white matter that can be computed from a simple formula relating the myelin volume fraction (MVF) to the fiber volume fraction (FVF). Recent studies have suggested that the sexual dimorphism in white matter development is due to a higher g-ratio (thinner myelin) in adolescent boys. Additionally, in vivo imaging of the myelin thickness in multiple sclerosis could provide a real-time tool for tracking myelination in lesions, facilitating the development and evaluation of new therapeutic agents that promote remyelination. In this abstract, a unique combination of magnetization transfer, diffusion imaging and histology is presented, providing a novel method for validating the in vivo measurements of the myelin g-ratio. We show that the g-ratio computed from MRI exhibits a high correlation with histology.

Silvia De Santis^1,2, Daniel Barazany^1,2, Derek K Jones¹, and Yaniv Assaf^2
1School of Psychology, CUBRIC, Cardiff, UK, United Kingdom, ²Department of Neurobiology, Tel Aviv University, Tel Aviv, Israel, Israel

The purpose of this work is to develop a new acquisition&analysis strategy, by combining inversion recovery with conventional diffusion tensor imaging, and by acquiring CHARMED protocol to calculate fiber orientations and volume fractions. For each fiber population present within a voxel, we extract a specific longitudinal relaxation time T1 by exploiting the orientational dependence of the diffusion-weighted signal that has been previously inversion-prepared. As the relaxation time T1 has been established as a solid proxy for myelination, this method effectively succeeds, for the first time, to resolve both axonal and myelin properties in presence of crossing fibers.

Maira Tariq¹, Torben Schneider², Daniel C Alexander¹, Claudia AM Wheeler-Kingshott², and Hui Zhang^1
1Department of Computer Science & Centre for Medical Image Computing, University College London, London, United Kingdom, ²NMR Research Unit, Department of Neuroinflammation, UCL Institute of Neurology, University College London, London, United Kingdom

We present a clinical technique to estimate the anisotropy of the orientations distribution of neurites, using Neurite Orientation Dispersion and Density Imaging (NODDI) technique. We show that NODDI can be utilised in vivo with a more realistic description of orientations of neurites, using a clinically feasible protocol, while the NODDI technique introduced in the original publication is still an accurate model for estimating neurite density and their concentration about the dominant orientation.

Manisha Aggarwal¹, Linda J Richards², and Susumu Mori^1
1Department of Radiology, Johns Hopkins University School of Medicine, Baltimore, MD, United States, ²The Queensland Brain Institute, University of Queensland, Brisbane, QLD, Australia

This work demonstrates diffusion MRI (dMRI) for imaging the microstructure of the developing cerebral cortex in the mouse embryo. 3D diffusion micro-imaging at high SNR was achieved using accelerated diffusion-weighted gradient and spin echo based acquisition. The resulting dMRI data allowed resolving the microscopic structure of transient zones in the developing cortex from embryonic day 12 (E12) to E18 (n=3 at each stage) based on diffusion as an endogenous probe, and revealed 3D imaging of cortical microstructure in unprecedented detail.

Corey A Baron¹, Mahesh P Kate², Laura C Gioia², Ken Butcher², Derek Emery³, Matthew D Budde⁴, and Christian Beaulieu^1
1Biomedical Engineering, University of Alberta, Edmonton, Alberta, Canada, ²Neurology, University of Alberta, Alberta, Canada, ³Radiology, University of Alberta, Alberta, Canada, ⁴Department of Neurosurgery, Medical College of Wisconsin, Wisconsin, United States

The mechanisms behind the marked reduction of mean diffusivity (MD) of water using standard pulsed gradient spin echo (PGSE) diffusion MRI after acute ischemic stroke are still not well understood. Here, oscillating gradient spin-echo (OGSE) diffusion MRI that enables short diffusion times of 4 ms demonstrated only a 14% drop of MD within human acute ischemic stroke lesions, as opposed to a 44% drop using PGSE with a much longer 40 ms diffusion time. This agreed well with Monte Carlo simulations of axon beading at the two diffusion times, supporting its role for MD reductions during stroke.