Neuron-based Measurements for Brain Functionality Understanding
Abstract
A two-photon imaging technique is able to capture a large number of in-vivo multi-region neuronal activities simultaneously. We carry out a causal interaction study on neuronal activities of miceÕs visual cortex. We applied a proposed data-driven causal interaction measure (earlier developed in our Laboratory) to the florescent calcium data. By estimating the fractal dimension of the delay-embedding point cloud, we give a nonlinear causality measure and a signal delay estimation between activities of an ordered pair of neurons. This leads to a weighted directed network of neurons or a binary network by thresholding. Despite high variability of single neuron activities over trials, for the same stimulus, we observed consistent patterns of the networks, such as hub neurons, population activities and topological characteristics. We defined and computed dissimilarity measure between networks to show our method is robust against trial-to-trial variability.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Oct 06, 2018
- Source ID
- W911NF1610395
Entities
People
- Hamid Krim
Organizations
- Army Contracting Command
- North Carolina State University
- United States Army