Epileptic Seizure Prediction from Eigen-Wavelet Multivariate Self-Similarity Analysis of Multi-Channel EEG Signals

Epileptic patients may suffer from severe brain damages during seizures. There is thus a significant need for automated seizure prediction. Independently, brain macroscopic activity has been shown to display scalefree temporal dynamics, which, in turn, were involved into seizure prediction. self-similarity, the paradigm model for scalefree dynamics, has however mostly been defined in univariate settings, thus yielding a collection of independent analyses of recorded signals. Yet, non-negligible correlations exist in multi-channel recordings of brain activity and may prove useful in seizure prediction. This work aims to assess the benefits of using a recently developed mul-tivariate eigen-wavelet framework for multivariate self-similarity analysis in seizure prediction using CHB-MIT Scalp EEG data.