Biogeography based hybrid scheme for automatic detection of epileptic seizures from EEG signatures

详细信息    查看全文

• 作者：Rohtash Dhiman^a ; ^{rohtash.k@gmail.com} ; ^{rohtash.ee@dcrustm.org}Author Vitae ; J.S. Saini^a ; ^{jssain@rediffmail.com}Author Vitae ; Priyanka^b ; ^{priyankaiit@yahoo.co.in}Author Vitae
• 关键词：Biogeography based optimization (BBO) ; Twin support vector machine (TWSVM) ; Electroencephalogram (EEG) ; Wavelet packet transform (WPT)
• 刊名：Applied Soft Computing
• 出版年：2017
• 出版时间：February 2017
• 年：2017
• 卷：51
• 期：Complete
• 页码：116-129
• 全文大小：1399 K
• 卷排序：51

文摘

EEG signals decomposed using Wavelet Packet Analysis, followed by computation of various statistical & Non-statistical properties to extract features. Biogeography based optimization is employed for selection of relevant features. The Habitat Suitability Index (HSI) values for each island in every population are computed by Twine Support Vector Machine. The Seizure detection rate for different classification instances achieved is 100%.