A neural network model for simulation of electrical stimulation
Abstract
Electrical stimulation of the brain is used for therapeutic and clinical diagnostic reasons. Modeling studies of neuronal response to electrical stimulation so far have dealt mostly with non-active neuronal networks and have focused on current distribution and voltage threshold effects for action potential induction in a single axon (myelinated or unmyelinated). This has created a huge gap in the understanding of the responses of single non-active neurons and active neurons in a network to an electric field. This gap partly results from the fact that the typical network simulator software packages are not designed to model neuronal networks that are subjected to electric fields. A detailed understanding of the interaction of electrical field with active and connected neurons would improve the efficacy of the existing therapies and devices and will help design new ones. This proposal aims to gain an understanding of how electrical stimulation affects the behavior of active neuronal networks. It focuses on creating and implementing new modeling techniques for simulations of large-scale biologically realistic cortical networks subjected to electrical stimulation. Traditional neuronal network modeling approaches assume that processing of inputs in neurons takes place predominantly in dendrites and therefore substitute neuronal axons for delay lines that introduce delay while inputs are transited between neurons. The electric field affects dendrites and axons and therefore such an approach cannot be used during the time course of applied electrical stimulation. This proposal will attempt to resolve the above problems by introducing to the modeled neurons all necessary components needed for modeling of the effects of applied electric fields on the level of cellular compartments in an active network. It requires special programming efforts that combine the complexity of the axonal tree and activation of postsynaptic neurons during the time course of stimulation. These new components include multicompartmental axons traversing through the volume of a simulated network and allowing action potentials propagate with a realistic speed as well as dedicated spike detector at each axonal terminal. The main innovation of this proposal is the newly designed neuronal model, which is equally suited to respond to synaptic input in both the absence and presence of an external electric field. Here we propose to study the effect of cortical stimulation in a model of an active cortical network exhibiting realistic or pathologic behaviors. The effect of electrical stimulation will be studied in a 3D model of cortex using multicompartmental neuronal representations designed to synaptically interact with other neurons in the presence and absence of an electric field. Our hypothesis is that the modulatory effect of electrical stimulation in cortical tissue is predominantly mediated by two factors impacting the network-level activity: 1) presynaptic action potentials generated in the distal and proximal axonal elements at the time when an electrical stimulation is applied; and 2) altered probability of neuronal firing resulting from current injection into the soma and the other neuronal compartments of neurons responding to electric stimulation. The project will seek to provide information in a quantitative manner on the effects of electrical stimulation on the activity in a simulated network as a function of the applied electric field. If successful, the results will significantly improve the efficacy of the existing electrical stimulation therapies across multiple application domains and would be useful for future device/therapies design.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jul 09, 2020
- Source ID
- W911NF2010183
Entities
People
- Paweł Kudela
Organizations
- Army Contracting Command
- Johns Hopkins University
- United States Army