Detecting Epileptic Seizures Using Electroencephalogram: A Novel Frequency Domain Feature Extraction Technique for Seizure Classification using Fast ANFIS
No Thumbnail Available
Date
2012
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Epileptic seizures usually results in a mixture of temporal alterations
in perception and behavior. Epilepsy is considered to be one of the
highly frequent neurological disorders. A considerable manner for
detecting and examining epileptic seizure behavior in humans is
Electroencephalogram (EEC) signal examination. EEC classification
is a significant process in Brain Computer Interface (BCl) that offers
a new dimension in human computer interface, directly linking a
computer with human thinking. Identification of the epileptic EEC
signal has been performed manually. Recently, automated epileptic
seizure identification with the help of EEC signals has become an
active of research. This paper presents an implementation of
automated epileptic EEG detection system. In this paper, frequency
domain feature extraction is carried out through Fast Fourier
Transform to the process of classifying EEG signals. For
classification this paper uses Fast Adaptive Ncuro-Fuzzy Inference
System (ANFIS) which utilize the modified Levenberg-Marquardt
algorithm for learning. Experimental results show that the proposed
system results in higher accuracy of classification at lesser time.