Title :
Prediction of hand movement from the event-related EEG using neural networks
Author :
Erfanian, Abbas ; Moghaddas, Saeid
Author_Institution :
Dept. of Biomed. Eng., Iran Univ. of Sci. & Technol., Tehran, Iran
Abstract :
This article explores the use of single-channel EEG to predict the hand movement including grasping, opening, and holding. A feedforward neural network with the backpropagation learning rule was trained to discriminate between three different patterns of EEG using the mean absolute value (MAV), variance, and the relative power of the Beta band to the Alpha band as the features. It was found that 80% of the novel data and 98.7% of the trained data were classified correctly
Keywords :
backpropagation; biomechanics; electroencephalography; feedforward neural nets; medical signal processing; EEG patterns; alpha band; backpropagation learning rule; beta band; electrodiagnostics; event-related EEG; grasping; hand movement prediction; holding; opening; trained data; Biological neural networks; Electrodes; Electroencephalography; Feedforward neural networks; Grasping; Intelligent systems; Mathematics; Multilayer perceptrons; Neural networks; Physics;
Conference_Titel :
[Engineering in Medicine and Biology, 1999. 21st Annual Conference and the 1999 Annual Fall Meetring of the Biomedical Engineering Society] BMES/EMBS Conference, 1999. Proceedings of the First Joint
Conference_Location :
Atlanta, GA
Print_ISBN :
0-7803-5674-8
DOI :
10.1109/IEMBS.1999.802508
