Automatic preprocessing of EEG signals in long time scale

Electroencephalography (EEG) signals are highly affected by physiological artifacts. Establishing a robust and repeatable EEG pre-processing is fundamental to overcome this issue and be able to use fully EEG data especially in long time scale experiments. In this work, starting from the Independent Component Analysis (ICA) of the EEG data, a control feedback scheme aiming to manage the cleaning of the independent component signals in an automatic way avoiding cut-bind solutions is presented, both with and without co-registrations. The method implemented combines different approaches based on the residual artifact contents check, identification and cleaning. The results of this procedure are shown on a test dataset. This analysis tool is embedded as core module, in a platform that can manage the automatic clearing of EEG recordings for multiple-subjects studies.