Evaluating classifiers to detect arm movement intention from EEG signals.

摘要: This paper presents a methodology to detect the intention make reaching movement with arm in healthy subjects before actually starts. is done by measuring brain activity through electroencephalographic (EEG) signals that are registered electrodes placed over scalp. The preparation and performance of an generate phenomenon called event-related desynchronization (ERD) mu beta frequency bands. A novel characterize this cognitive process based on three sums power spectral frequencies involved ERD presented. main objective set benchmark for classifiers choose most convenient. best results obtained using SVM classifier around 72% accuracy. will be used further research control commands move robotic exoskeleton helps people suffering from motor disabilities perform movement. final aim brain-controlled improves current rehabilitation processes disabled people.