This work investigates the hypothesis that focal seizures can be predicted using scalp electroencephalogram (EEG) data. Using deep learning, we extract representations that distinguishes between EEG activity relevant to seizure prediction. Convolutional filters on the wavelet transformation of the EEG signal are used to define and learn quantitative signatures of three types of EEG activity: nonseizure, preseizure, and seizure. The optimal seizure prediction horizon is also learned from the data as opposed to making an a priori assumption. In our results, computational solutions to the optimization problem indicate a ten-minute seizure prediction horizon. This result is verified by measuring Kullback-Leibler divergence on the distributions of the automatically extracted features. The results on the EEG database of 204 recordings demonstrate that (i) the preseizure phase transition occurs approximately ten minutes before seizure onset, and (ii) the prediction results on the test set are promising, with a sensitivity of 87.8\% and a low false prediction rate of 0.142 FP/h. Our results significantly outperform a random predictor and other seizure prediction algorithms. We demonstrate that a robust set of features can be learned from scalp EEG that characterize the preseizure state of focal seizures.
Building 9 - Lecture hall 2
14:30 - 14:45