The following paper proposing a novel epileptic seizure detection method based on EEG has been accepted by IEEE Access.
Akira Furui, Ryota Onishi, Tomoyuki Akiyama, and Toshio Tsuji
Epileptic seizure detection using a recurrent neural network with temporal features derived from a scale mixture EEG model
IEEE Access (accepted).
This research extends the master's thesis work of Mr. Ryota Onishi, an alumnus of the Biological Systems Engineering Laboratory at the Faculty of Engineering. We demonstrated that epileptic seizures can be detected with high accuracy by evaluating EEG amplitude and non-Gaussianity using a probabilistic model while considering temporal dynamics with an RNN.