Detection of Epileptic Seizure using Radius Measurement and Higher Order Moments in The EMD Domain
Kaushik Das1, Rajkishur Mudoi2
1Kaushik Das, Department of Electronics and Communication Engineering, North-Eastern Hill University, Shillong (Meghalaya), India.
2Rajkishur mudoi, Department of Electronics and Communication Engineering, North-Eastern Hill University, Shillong (Meghalaya), India.
Manuscript received on 13 April 2017 | Revised Manuscript received on 20 April 2017 | Manuscript Published on 30 April 2017 | PP: 170-175 | Volume-6 Issue-4, April 2017 | Retrieval Number: D4957046417/17©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© 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: This paper presents a method for the detection of epileptic seizure from EEG signal using empirical mode decomposition (EMD). The intrinsic mode functions (IMFs) which is generated by the EMD method can be considered as a set of amplitude and frequency modulated (AM-FM) signals. The Hilbert transformations of these IMFs which is circular form in the complex plane can be used as a feature for radius calculation and the higher order moments like variance, skewness and kurtosis are applied on the output values of Short-time Fourier transform (STFT) of the IMFs, the proposed method shows better classification result than simply applying higher order moments. The effectiveness of the proposed method is tested using the dataset which is available online. It is found that the result obtained from radius measurement and higher order statistical moments provide good discrimination performance for the detection of epileptic seizure.
Keywords: Electroencephalogram (EEG), Intrinsic Mode Functions (IMFs), Empirical Mode Decomposition (EMD), Epileptic Seizure.
Scope of the Article: Measurement & Performance Analysis