Phenomenological Mesoscopic Models for Seizure Activity
In this chapter we review phenomenological models of seizure like activity. We discuss dynamical mechanisms for seizure onset and offset, preictal spikes, spike and wave complexes and status epilepticus, highlighting the role played by the bifurcation structure of the model, the presence of noise and the emergence of multiple interacting time-scales. These models can be used to build large-scale patient specific brain network models serving as in-silico platforms to test clinical hypothesis and perform virtual surgeries. They suggest innovative treatment strategies, such as minimally invasive ablations or stimulations that fully exploit the network and dynamical properties of the system, or even modulation of variables and parameters to force the system in safer regions of the bifurcation diagram. We discuss insights from phenomenological models that can help to foster our understanding of the mechanisms underlying epileptic seizures.