Recursive Model Identification for the Analysis of Cardiovascular Autonomic Modulation During Epileptic Seizure

Abstract

Significant cardio-respiratory fluctuations are often observed during and after an epileptic seizure event. The mechanisms underlying these acute modifications are considered to be involved in sudden and unexpected death in epilepsy (SUDEP). We hypothesize that these acute events are mediated by specific dynamics of the autonomic nervous system (ANS). However, the evaluation of the ANS during seizures remains particularly challenging, mainly due to the lack of observability. Computational modelling could help to override these limitations, to assess ANS modulation and to evaluate this hypothesis. In this study, we propose and apply a recursive identification algorithm of a system-level model of the autonomic modulation of the sino-atrial node, integrating a Tikhonov regularization, in order to assess sympathetic and parasympathetic activities during ictal tachy-bradycardia events. We evaluate the feasibility of the method on heart rate (HR) data from 4 seizures observed in the same patient. After parameter optimization and identification we were able to reproduce observed HR data with a maximum root mean squared error equals to 1.7bpm. The estimated autonomic series show sympathetic activation and parasympathetic inhibition at the seizure onset, and a massive vagal discharge as the leading factor to ictal bradycardia.