Analysis and classification of sleep EEG | ||||
| The International Conference on Electrical Engineering | ||||
| Article 79, Volume 8, 8th International Conference on Electrical Engineering ICEENG 2012, May 2012, Page 1-8 PDF (94.26 K) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/iceeng.2012.32692 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
| Noha E. El-Kafrawy1; F. E. Z. Abou-Chadi2; S. I. Rihan3 | ||||
| 1Egyptian Armed Forces. | ||||
| 2Benha High Technology Institute, Benha, Egypt. | ||||
| 3College of Engineering, Cairo University, Cairo, Egypt. | ||||
| Abstract | ||||
| In the present paper, a comparative study of performance for three techniques of feature extraction is presented in order to classify the sleep stages using EEG signals. A multilayer feed forward neural network was used for classification. Six sleep EEG records for each of ten patients were selected from Cairo Center of Sleep Disorder. Three methodologies of analysis were utilized for feature extraction. These include: autoregressive modeling (AR), bispectral analysis, and discrete wavelet transform (DWT), where principle component analysis (PCA) was used to reduce feature dimensionality. The features derived from the three methodologies of signal analysis were used as input feature vectors to the classifier. Information fusion is very important task in pattern recognition as it is difficult to develop classifiers with a high identification performance rate. The multilayer feed forward neural network gives higher classification rate using the data fusion at the feature extraction level. It reaches 83.4%. | ||||
| Keywords | ||||
| Autorgressive modeling; Bispectral Analysis; Discrete Wavelet Transform; principle componenet analysis; multilayer feed forward neural network | ||||
|
Statistics Article View: 93 PDF Download: 209 |
||||
