Home Computer Science Computational Diffusion MRI: MICCAI Workshop, Athens, Greece, October 2016
Colocalization of Functional Activity and Neurite Density Within Cortical Areas
Achille Teillac, Sandrine Lefrance, Edouard Duchesnay, Fabrice Poupon, Maite Alaitz Ripoll Fuster, Denis Le Bihan, Jean-Francois Mangin, and Cyril Poupon
Abstract In this work, we investigated the link between the blood-oxygen-level dependant (BOLD) effect observed using functional magnetic resonance imaging (fMRI) and the neurite density inferred from the Neurite Orientation Dispersion and Density Imaging (NODDI) model in some well-known lateralized cortical areas. We found a strong colocalization between those two parameters in lateralized areas such as the primary motor cortex, the language network, but also the primary visual cortex, which might indicate a strong link between microstructure and functional activity.
Since the early twentieth century, neuroanatomists aim at disentangling the link between the functional organization of the brain and the tissue microstructure at the cellular scale. Brodmann was among the first to publish his cytoarchitectonic atlas built from the observation of cortical histological slices using optical microscopy, that is debated but still remains used today. Its construction relying on very few subjects is a strong limitation since it prevents from investigating the intersubject variability and gives limited confidence about the boundaries separating the established cortical areas. In addition, it prevents from any comparison with brain functioning. Recently, the emergence of diffusion-based MR microscopy providing quantitative features of the tissue microstructure such as the axon diameter and density [1, 2] may open an avenue to the establishment of novel atlas of the brain cytoarchitecture. It relies on in-vivo acquisition and therefore enables to look at the variability of these features between subjects and to correlate them with brain functions investigated using functional MRI. Moreover, the access to such
S. Lefrance • E. Duchesnay • F. Poupon • J.-F. Mangin NeuroSpin/UNATI, I2BM, CEA, Gif-sur-Yvette, France
© Springer International Publishing AG 2017
A. Fuster et al. (eds.), Computational Diffusion MRI, Mathematics
and Visualization, DOI 10.1007/978-3-319-54130-3_15
quantitative features of the ultrastructure may bring valuable imaging markers of brain diseases since most pathological mechanisms occur at the cellular level.
For this study, the NODDI model  has been chosen because of its relatively easy use in clinical routine involving a multiple-shell diffusion-weighted MRI protocol and because the underlying modeling of the tissue relies on the definition of three compartments: a compartment corresponding to the CSF characterized by a fast isotropic diffusion process mainly used to account for partial volume effects, a restricted cylinder compartment corresponding to the part of water constrained within the intra-dendritic and intra-axonal spaces represented by sticks, an hindered compartment corresponding to the extracellular space characterized by a Gaussian diffusion model. This extracellular model might not reflect the anatomical complex organization because of the Gaussian assumption for all the components in the extracellular compartment. However, the NODDI model has proven to be a useful tool to explore the cortex in some pathological diseases such as the focal cortical dysplasia as detailed in . It comes with several metrics of interest such as the intracellular fraction (fjntra) representing the local neurite density or the orientation dispersion index (ODI) being very small in white matter where fibers present a high degree of alignment and closer to 1 in the cortical ribbon where dendrites are almost arbitrarily oriented in space.
In this work, we investigate the relation between microarchitectural features measured using the NODDI model and the underlying functional activities observed using fMRI. The Materials and methods section describes the acquisition protocol, the preprocessing and computation of the NODDI maps, the pipeline to obtain the intracellular fraction and how it has been correlated with the z-score maps stemming from the fMRI; the Results section presents the obtained maps and describes more explicitly the relationship between the neurite density and the functional activity for cortical regions; the Discussion section summarizes the contribution and presents the potential of the tools addressed in this paper.
|< Prev||CONTENTS||Next >|