UKF-based slow-variable control for firing patterns in CA3 neurons

Epilepsy is commonly considered to be the result of excessive firing of neurons. This paper investigates firing patterns of single-neuron in the hippocampus CA3 area based on the Pinsky-Rinzel (PR) model and uses closed-loop control with slow variable as the feedback parameter to suppress neuronal firing. Unscented Kalman Filter (UKF) is adopted to estimate the slow variables which contain important information of the system from membrane potential. In this paper, we propose the basic control method: linear proportion-integration (PI) control to confirm the effectiveness of this strategy. The simulation results prove that this closed-loop can control neuronal firing patterns from firing to rest, and it may provide a basis for new and more effective therapies for people with epilepsy.