Improvement in the brain activity monitoring using EEG Signal Analysis and Convolution al neural network

Abstract

Electroencephalography is a method to analyze electrical activitiespresent in the different parts of the human brain and using visual trace it records these activities.EEG provides cost effective, portable, high frequency and accurate measurement as compare to other brain wave activity monitoring tool. The electrodes present in EEG test detects tiny electrical charges that result from the activity of brain cells. The interpretation of this large EEG signal provides increased accuracy for analyzing brain functionality. Traditional method of analysis of EEG signal relies more on the trained experts. In the proposed research using machine learning techniques and spatial temporal data the classification task of EEG signal analysis is performed with more accuracy and in time efficient manner. In this research sliding window protocol with specified window size is used for generation of training data for deep learning network and to standardize the training data for enhancement of model’s performance. For the accuracy comparison variantsofrecurrentneuralnetwork and Convolutionalneuralnetwork is analyzed using the collected dataset. Result analysis shows that the convolutional neural network produces high accuracy and training efficiency when compare with different machine learning model.