EPILEPTIC SEIZURES CAN BE ANTICIPATED BY NONLINEAR-ANALYSIS

Epileptic seizures are a principal brain dysfunction with important pu blic health implications, as they affect 0.8% of humans. Many of these patients (20%) are resistant to treatment with drugs'. The ability to anticipate the onset of seizures in such cases would permit clinical interventions. The view of chronic focal epilepsy now is that abnormal ly discharging neurons act as pacemakers to recruit and entrain other normal neurons by loss of inhibition and synchronization into a critic al mass(2). Thus, preictal changes should be detectable during the sta ges of recruitment. Traditional signal analyses, such as the count of focal spike density(3), the frequency coherence(4) or spectral analyse s are not reliable predictors. Non-linear indicators may undergo consi stent changes around seizure onset(5-7). Our objective was to follow t he transition into seizure by reconstructing intracranial recordings i n implanted patients as trajectories in a phase space and then introdu ce non-linear indicators to characterize them(8,9). These indicators t ake into account the extended spatio-temporal nature of the epileptic recruitment processes(10) and the corresponding physiological events g overned by short-term causalities in the time series. We demonstrate t hat in most cases (17 of 19), seizure onset could be anticipated well in advance (between 2-6 minutes beforehand), and that all subjects see med to share a similar 'route' towards seizure.