Seizure detection with a biologically plausible algorithm. We used Adaptive Spiking Neural Networks with Liquid-Time Constant Spiking Neurons and Forward Propagation Through Time (FPTT) for the detection of seizures in scalp-EEG datasets.

Code can be updated according to desired and relevant datasets. Code inspired by the research of

Accurate online training of dynamical spiking neural networks through Forward Propagation Through Time Yin, B., Corradi, F., & Bohté, S. M. (2023). Accurate online training of dynamical spiking neural networks through Forward Propagation Through Time. Nature Machine Intelligence, 1-10.

Paper title: "Biological Plausible Algorithm for Seizure Detection: Towards AI-enabled Electroceuticals at the Edge."