EEG mobility artifact removal for ambulatory epileptic seizure prediction applications

•Mobility artifacts affect the quality of EEG signals and mask seizure markers what reduces the classification accuracy of epileptic seizure detection and prediction algorithms.•Time and frequency domain characterization of mobility artifacts show that they affect the entire spectrum of EEG electrodes from all over the scalp.•Independent Component Analysis (ICA) is utilized as a blind source separation technique for removing mobility artifacts efficiently.•Seizure detection/prediction with ICA-based artifact removal has shown an average sensitivity increase of 8