Dynamic Entrainment: A deep learning and data-driven process approach for synchronization in the Hodgkin-Huxley model

Abstract

Resonance and synchronized rhythm are important phenomena in the nature and they could be either constructive or destructive in dynamical systems specifically in biology. There are many examples related to the fact that the human's body organs must maintain their rhythm in order to function properly. For instance, in the brain, desynchronized electrical activities could lead to neurodegenerative disorders like Huntington's disease. In this paper, we employ a well known conductance based neuronal model known as Hodgkin-Huxley model describing the propagation of action potentials in neurons. Armed with the data-driven process alongside the results of the Hodgkin-Huxley model, we introduce a novel Dynamic Entrainment technique, which is able to maintain the system to be in its entrainment regime dynamically by applying deep learning approaches.