Title :
Neural networks for EEG signal decomposition and classification
Author :
Vaz, Francisco ; Príncipe, José Carlos
Author_Institution :
Dept. de Electron. e Telecoms, Aveiro Univ., Portugal
Abstract :
The authors present a new method for the decomposition of time varying signals based on a principal component analysis implemented with unsupervised neural networks. The method is applied to EEG spike detection using neural networks directly fed with the signal components, avoiding the time consuming feature extraction phase. Results show good learning capability of the topologies used
Keywords :
electroencephalography; medical signal processing; neural nets; unsupervised learning; EEG signal classification; EEG signal decomposition; EEG spike detection; electrodiagnostics; epilepsy; learning capability; principal component analysis; signal components; time consuming feature extraction phase; time varying signals; Biological neural networks; Computer networks; Electroencephalography; Epilepsy; Feature extraction; Neural engineering; Neural networks; Neurons; Signal resolution; Telecommunication computing;
Conference_Titel :
Engineering in Medicine and Biology Society, 1995., IEEE 17th Annual Conference
Conference_Location :
Montreal, Que.
Print_ISBN :
0-7803-2475-7
DOI :
10.1109/IEMBS.1995.575366
