Classification of EEG using PCA, ICA and Neural Network
Kavita Mahajan1, M. R. Vargantwar2, Sangita M. Rajput3
1Mrs. Kavita Mahajan, Department of Electronics & Communication Engineering, Marathwada Institute of Technology, Aurangabad, India.
2Mrs. M. R. Vargantwar, Department of Electronics & Communication Engineering, Marathwada Institute of Technology, Aurangabad, India.
3Mrs. Sangita M. Rajput, Department of Electronics & Communication Engineering, S.S.V.P.S B.S.Deore College of Engineering, Dhule, India.
Manuscript received on October 06, 2011. | Revised Manuscript received on October 12, 2011. | Manuscript published on October 30, 2011 . | PP: 80-83 | Volume-1 Issue-1, October 2011. | Retrieval Number: A0117101111/2011©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The processing and analysis of Electroencephalogram (EEG) within a proposed framework has been carried out with DWT for decomposition of the signal into its frequency sub-bands and a set of statistical features was extracted from the sub-bands to represent the distribution of wavelet coefficients. Reduction of the dimension of the data is done with the help of Principal component analysis and Independent components analysis. Then these features were used as an input to a neural network for classification of the data as normal or otherwise. The performance of classification process due to different methods is presented and compared to show the excellent of classification process. These findings are presented as an example of a method for training, and testing a normal and abnormal prediction method on data from individual petit mal epileptic patients.
Keywords: ANN. DWT, Electroencephalogram (EEG), Independent components analysis (ICA), Principal component analysis (PCA),