Seizure intervention via diffusion mechanisms

Abstract

We begin by demonstrating that the neuronal state equation from Dynamic Causal Modelling takes on the form of the discretized Fokker-Planck equation upon the inclusion of local activity gradients within a network. Using the Jacobian of this modified neuronal state equation, we show that an initially unstable system can be rendered stable via a decrease in diffusivity. Therefore, if a node within the system is known to be unstable due to a tendency toward epileptic seizures for example, then our results indicate that intervention techniques should take the counter-intuitive approach of mirroring activity in the nodes immediately surrounding a pathological region - effectively fighting seizures with seizures. Finally, using a two-region system as an example, we show that one can also tune diffusivity in such a way as to allow for the suppression of oscillatory activity during epileptic seizures.