Inferring White Matter Structure from Correlations in Neural Population Activity

RABIYA Noori (1,2), JEREMIE Lefebvre (1,2,3)

  1. Fundamental Neurobiology, Krembil Research Institute

  2. Institute of Biomaterials and Biomedical Engineering, University of Toronto

  3. Department of Mathematics, University of Toronto

White matter (WM) plasticity plays an integral role in maintaining synchrony of brain network components by regulating conduction velocity in the propagation of neural signals. To explore the role of WM, we developed a brain-scale network model to simulate and isolate the effect of myelin structure on synchronization patterns in the human brain. This model is built of interacting oscillatory nodes, each representing an individual brain region, and described by a system of first-order nonlinear stochastic delay differential equations. Given population dynamics data for regions across the whole brain, we can optimize our model to find the set of parameters that will give rise to the given target dynamics. We have applied this approach to estimate network weights and also isolate conduction velocities along myelin tracts in our model, to get an indirect measure of myelination. We are aiming to apply our approach on real MEG measurements from two cohorts: healthy vs patients with WM degradation, and are expecting to see significant difference in how the distribution of signal delays across the brain networks differ. We also examined how robust our results were as a function of the oscillation frequency of the neural populations involved. By learning network structure using neural dynamics data we will be able to leverage our knowledge of the biophysical representation of the parameters and structure of the model, to get insight on the role of myelin in brain dynamics. Our approach will improve our understanding of disorders that affect WM plasticity, such as multiple sclerosis.Rab

Neural SystemsiARC