Optimal control based seizure abatement using patient derived connectivity.

Peter N Taylor,Justin Ruths,Marcus Kaiser,Thomas Thesen,Justin Dauwels,Jijju Thomas,Nishant Sinha

doi:10.3389/fnins.2015.00202

Abstract

Epilepsy is a neurological disorder in which patients have recurrent seizures. Seizures occur in conjunction with abnormal electrical brain activity which can be recorded by the electroencephalogram (EEG). Often, this abnormal brain activity consists of high amplitude regular spike-wave oscillations as opposed to low amplitude irregular oscillations in the non-seizure state. Active brain stimulation has been proposed as a method to terminate seizures prematurely, however, a general and widely-applicable approach to optimal stimulation protocols is still lacking. In this study we use a computational model of epileptic spike-wave dynamics to evaluate the effectiveness of a pseudospectral method to simulated seizure abatement. We incorporate brain connectivity derived from magnetic resonance imaging of a subject with idiopathic generalized epilepsy. We find that the pseudospectral method can successfully generate time-varying stimuli that abate simulated seizures, even when including heterogeneous patient specific brain connectivity. The strength of the stimulus required varies in different brain areas. Our results suggest that seizure abatement, modeled as an optimal control problem and solved with the pseudospectral method, offers an attractive approach to treatment for in vivo stimulation techniques. Further, if optimal brain stimulation protocols are to be experimentally successful, then the heterogeneity of cortical connectivity should be accounted for in the development of those protocols and thus more spatially localized solutions may be preferable.

Highlights

Epilepsy is a spectrum of disorders in which patients have seizures due to abnormal neuronal activity (Blumenfeld and Taylor, 2003)
Optimal control of epileptic spike-wave dynamics often associated with absence seizures, myoclonic seizures and complex partial seizures (Asconapé and Penry, 1984; Sadleir et al, 2006)
Imaging Cortical connectivity was inferred from a 22 year old female patient clinically diagnosed with idiopathic generalized epilepsy with a history of absence and generalized tonic clonic seizures

Summary

Introduction

Epilepsy is a spectrum of disorders in which patients have seizures due to abnormal neuronal activity (Blumenfeld and Taylor, 2003) Symptomatic manifestations of these events can include a loss of consciousness, tonic-clonic convulsions and myoclonic jerks, amongst others which can severely impact patient quality of life. These transient seizure events often have distinctive electrographic correlates detectable on the electroencephalogram (EEG). SWDs are high amplitude periodic oscillations with a frequency typically slower that of normal awake EEG They are Optimal control of epileptic spike-wave dynamics often associated with absence seizures, myoclonic seizures and complex partial seizures (Asconapé and Penry, 1984; Sadleir et al, 2006). The first line of treatment for patients with epilepsy is typically medication, in over 30% of cases medication alone is insufficient (Keränen et al, 1988)

Methods

Results

Conclusion