Abstract

The processes by which the brain switches from normal activity to an epileptic seizure have major implications for seizure prevention and treatment, yet remain largely unknown. Seizure onset can be described as a critical transition (CT), but there is no consensus whether (i) bifurcation-induced, (ii) noise-induced, or (iii) bifurcation/noise-induced CTs are responsible. To clarify this, we develop a versatile CT-classification framework that can be applied to seizures in both animals and humans. First, we identify a canonical mathematical model which displays CTs that closely resemble voltage recordings of real seizures and can be of the three types mentioned above. We then identify distinctive properties of each CT type in the model’s output and use them to train a machine learning CT-type classifier. Finally, we apply the model-trained classifier to voltage recordings from epileptic rodents. We find that the largest proportion of analysed seizures are classified as noise-induced CTs. This challenges the conventional view that seizures are predominantly bifurcation-induced and could inform seizure prevention and treatment strategies.    Joint work with: Cian McCafferty (University College Cork), Klaus Lehnertz (University of Bonn Medical Centre, University of Bonn), François David (Collège de France), Vincenzo Crunelli (University of Lisbon), William P. Marnane (University College Cork), and Sebastian Wieczorek (University College Cork).