Ahmad Chamseddine1, Mohamad Sawan2
15:30 - 17:00 | Mon 29 Oct | Ambassador C | A4L-A.3
Seizure prediction has been a great challenge for neuroscientists in the last decade. Forecasting an epileptic episode has a substantial role in preventing or mitigating the harm that comes along, especially in some medically intractable cases. Despite the remarkable breakthrough in Artificial Intelligence and automated reasoning, the prediction of the epileptic seizure is still challenging due to the lack of dataset. Additionally, EEG signal are patient-specific, thus making it difficult to benchmark findings. In this study, instead of classifying short EEG segment, we detect the preictal phase as one long sequence of a latent space representation. Using Long Short-Term Memory (LSTM) neural network helps to obtain promising results. In fact, we proved that LSTM could be applied successfully as a post-processing regularizer for the classifier output.