Information Flow in the White Matter During a Motor Task: A Structural Connectivity Driven Approach

Guillermo Gallardo (1, 2) , Demian Wassermann (2, 1) , Rachid Deriche (2, 1) , Maxime Descoteaux (3) , Samuel Deslauriers-Gauthier (2, 1)

1 ATHENA - Computational Imaging of the Central Nervous System
2 COMUE UCA - COMUE Université Côte d'Azur (2015-2019)
3 SCIL - Sherbrooke Connectivity Imaging Lab [Sherbrooke]

Cognitive tasks emerge from the interaction of functionally specialized cortical regions (Verhagen et al. 2013). These interactions are supported by information flow through white matter fiber bundles connecting distant cortical regions. Estimating the information flow through white matter fiber bundles would therefore provide valuable information into the necessary cortical interactions to realize a task. In this work, we build a Bayesian network representing cortical regions and their connections using a structural connectivity driven parcellation (Gallardo et al., 2016) derived from diffusion MRI (dMRI). We then introduce Magnetoencephalography (MEG) measurements as evidence into this network to infer the information flow between cortical regions (Deslauriers-Gauthier et al., 2016). We show, for the first time, results on the interaction between the precentral, postcentral and occipital regions during a hand-movement task.

Domaines: Imagerie médicale

Guillermo Gallardo, Demian Wassermann, Rachid Deriche, Maxime Descoteaux, Samuel Deslauriers-Gauthier. Information Flow in the White Matter During a Motor Task: A Structural Connectivity Driven Approach. OHBM 2017 Organization for Human Brain Mapping Annual Meeting, Jun 2017, Vancouver, Canada. ⟨hal-01534978⟩