ICA and a gauge of filter for the automatic filtering of an EEG signal

The EEG signal, a recording of the brain activity using multiple electrodes placed on the scalp, can be hardly contaminated by a lot of noises called artifacts. The artifacts are generated by the action of the skeletal muscles such as: eye movements, jaw clenching, etc. Indeed, the signals recorded are mixture of phenomenon from multiple generators. Therefore it is important to search for a method which can separate the muscular activity from the neuronal one. The ICA is a statistical analysis method largely used for the study of biomedical data. Using the data recorded from several subjects (epileptic and healthy) we are going to prove the effectiveness of our approach based on the ICA and on a characterization of a filter model.