Description

A Deep Transfer Learning Approach for Seizure Detection Using RGB Features of Epileptic Electroencephalogram Signals

Abstract:

                    In this project,based on Deep transfer learning approach for seizure detection of Epileptic Electroencephalogram signals.However, the Epileptic seizure prediction and classification performance is not satisfactory over small EEG dataset using traditional approaches. The Transfer learning approach overcomes this by reusing the pre-trained networks such as googlenet, resnet101 and vgg19 trained on large Image database. This experiment has been done in two phases: (1) RGB image dataset generated for the seizure and non-seizure EEG signals data of University of Bonn using a novel preprocessing technique, (2) we configured googlenet, resnet101 and vgg19 trained networks to learn a new pattern or features from the RGB image Dataset and finally, above mentioned networks have been used for the classification.We will mainly emphasize on the results obtained from the googlenet, since it provides effective accuracy taking less time for prediction.

Introduction:

               World Health Organization (WHO) reported in that Epilepsy affecting about 50 million people worldwide of the world’s population.In where, EEG plays a significant role in epilepsy diagnosis.EEG report prepared by sensing electrical activities within brain neurons with respect to time using electrodes over the scalp and thus provides spatial and temporal information of the brain. The brain contains billions of cells, half of them are neurons and the other half facilitates the activity of neurons. The brain cells or neurons communicate using electrical impulses.EEG contains massive useful information about the highly random and complex neural activities. Due to random and complex neural activities, EEG signals are non-linear and non-stationary in nature. It is pretty problematic to perform diagnosis through simply observing the signals in the time domain. Therefore, important features of EEG signals are required for the diagnosis of different neurological diseases.

Proposed system:

              In this paper,  seizure detection using RGB features of Epileptic Electroencephalogram signals using Deep learning techniques.The proposed system is trained with image datasets.The network is created using googlenet with transfer learning techniques.RGB images have been generated for the both seizure and seizure free EEG signals