Call For Paper Volume:4 Issue:10 Oct'2017 |

Classification of EEG Signals using Empirical Mode Decomposition and Neural Networks

Publication Date : 24/06/2015



Author(s) :

Santosh Hiremani , D. A. Torse , N. J. Inamdar.


Volume/Issue :
Volume 2
,
Issue 6
(06 - 2015)



Abstract :

The brain is the central part of the human nervous system and is highly complex organ. The problems related to brain activities are analyzed and the diagnosis is carried out with the help of classification of normal and abnormal signals produced by the brain activity. Several methods were proposed for the classification of the EEG normal and abnormal waves which are non stationary waves. The classification of seizure (abnormal) and non-seizure (normal) signals is very important in the diagnosis of problems related to brain and some physiological disorders. This paper presents a method to classify the EEG signals with the help of Empirical mode decomposition which produces intrinsic mode functions (IMFs) and they can be considered as amplitude and frequency modulated (AM-FM) signals. The Hilbert transformation of IMFs helps to represent the signals in analytical form. The two bandwidths, namely amplitude modulated bandwidth and frequency modulated bandwidth calculated from the analytic IMFs, have been used as input to artificial neural networks (ANN) for classifying normal and abnormal EEG signals. Available EEG signals are used to show the effectiveness of the proposed method.  


No. of Downloads :

3


Indexing

Web Design MymensinghPremium WordPress ThemesWeb Development

Classification of EEG Signals using Empirical Mode Decomposition and Neural Networks

June 19, 2015