Skip to main content
Fig. 2 | BioMedical Engineering OnLine

Fig. 2

From: A knowledge discovery methodology from EEG data for cyclic alternating pattern detection

Fig. 2

Schematization of the feature computation process based on the filtered EEG signal. In the boxes representing the segmentation process, aiming to represent the percentage of overlap, full lines describe a given window at discrete time k, and dashed lines represent the same window but at time k + 1. “FE” boxes represent feature extraction, and the “FC” box represent the computation of the \(MMSD_{\varphi }\) feature that result from the combination of \(C_{\varphi , \tau }\) and \(C_{{\varphi ,\tau_{0} }}\)

Back to article page