An Optimal Compressive Approach for the Analysis of Electroencephalogram using Artificial neural network

Abstract

Electroencephalogram is an important tool for diagnosing, Monitoring and Managing neurological disorders related to brain diseases
such as epilepsy, tumor, encephalitis, and sleep disorders; using hardware like Electroencephalogram (EEG) wherein the strokes are directly
detected and the brain signals is recognized. The Proposed work is an algorithm to perform digital data retrieval in a denoised signal and to
design a vector quantizer for digital signal compression using clustering and subvector technique. A challenge of brain signal denoising is how to
preserve the edges of brain signal when reducing noise. The paper presents an approach for signal denoising based on wavelets Thresholding.
The presence of brain activity in the Electroencephalogram (EEG) confirms the diagnosis of diseases. The purpose of the work describes the
automated detection of brain diseases based on wavelet analysis of electroencephalogram. Three layer feed forward back-propagation artificial
neural network (ANN) is designed to classify the brain signals among different age group of people. The algorithm is developed based on the
lossless digital data retrieval concept.

Keywords: Electroencephalogram (EEG), Artificial Neural Networks (ANN), watermarking, wavelet Thresholding, Vector quantization.