Utilizing Empirical Mode Decomposition and Deep Learning Procedures for The Detection and Analysis of Epilepsy Disorders

In this the establish and diagnose of the neural illness called epilepsy with the application of the deep learning approach. This is achieved by making a clear distinction between the focal EEG information and the nonfocal EEG data. In relation to this particular study endeavour, the process of identifying epilepsy sickness through the implementation of soft computing techniques is described. The components of the suggested method are feature extraction and classification that utilizes decomposition in time-scale signal as well as diagnosis. To achieve the EMD, the electroencephalogram (EEG) signal is decomposed into 6 IMF sub-bands. Each of the calculated exterior fundamental topographies of the deconstructed IMF sub bands is then fed through the suggested Convolutional Neural Networks (CNN) architecture together with the quantities of each IMF sub band to be accomplished and more confidential. This process is done continuously until the right classification is made is made. It is worth stressing that the CNN classifier is used to solve the problem of focal or non-focal classification of the collected electroencephalogram (EEG) data. Last but not the least, the focal indication is categorized as moderate situation or severe case based on the strategy suggested in this research using CNN architecture.