An Efficient Approach to Define the Input Stimuli to Suppress Epileptic Seizures Described by the Epileptor Model

Nenhuma Miniatura disponível




Brogin, João Angelo Ferres [UNESP]
Faber, Jean
Bueno, Douglas Domingues [UNESP]

Título da Revista

ISSN da Revista

Título de Volume



Epilepsy affects about 70 million people in the world. Every year, approximately 2.4 million people are diagnosed with epilepsy, two-thirds of them will not know the etiology of their disease, and 1% of these individuals will decease as a consequence of it. Due to the inherent complexity of predicting and explaining it, the mathematical model Epileptor was recently developed to reproduce seizure-like events, also providing insights to improve the understanding of the neural dynamics in the interictal and ictal periods, although the physics behind each parameter and variable of the model is not fully established in the literature. This paper introduces an approach to design a feedback-based controller for suppressing epileptic seizures described by Epileptor. Our work establishes how the nonlinear dynamics of this disorder can be written in terms of a combination of linear sub-models employing an exact solution. Additionally, we show how a feedback control gain can be computed to suppress seizures, as well as how specific shapes applied as input stimuli for this purpose can be obtained. The practical application of the approach is discussed and the results show that the proposed technique is promising for developing controllers in this field.



Epileptic seizure suppression, Epileptor model, fuzzy Takagi-Sugeno modeling, linear matrix inequalities

Como citar

International Journal of Neural Systems, v. 30, n. 11, 2020.