Figure 1
From: Unsupervised EEG preictal interval identification in patients with drug-resistant epilepsy

Block diagram of the proposed methodology. The study’s first phase corresponded to preprocessing and feature engineering and preparation. The second phase encompassed dimensionality reduction of each feature group (univariate linear, univariate nonlinear and multivariate), followed by the application of unsupervised learning methods and the preictal interval inspection.