Early prediction of epileptic attacks using Alexnet neuralnetwork

Predicting epileptic attacks before they occur can be effective in preventing them throughtherapeutic intervention. Electroencephalogram (EEG) signals are used to diagnose epilepticseizures. However, the screening system cannot accurately detect epileptic seizure states. Inthe proposed method, in order to predict epileptic attacks before seizures based onelectroencephalogram signals, a series of preprocessings, artificial neural networks and EEGdeep learning have been used to prevent the occurrence of epileptic attacks. First, the CHBMITScalp EEG dataset is used for experiments, and the EEG signals are divided into ۴ periodsof ۱۰ minutes after preprocessing. Then, for pre-processing, a notch filter is used to reducenoise, the inherent components of EMD and wavelet transform are used to analyze EEG signalsinto different sub-bands and extract features. Then these features are given to the support vectormachine (SVM) in order to classify the signal of the healthy and affected person, and the signalof the affected person is considered in two states, ictal and preictal, and finally, we use theconvolutional neural network (CNN) with the AlexNet architecture in MATLAB program bycomparing the features to predict epileptic attacks. The architectural sensitivity in predictingepileptic attacks is ۹۹%, the rate of wrong prediction of epileptic attacks is on average ۰.۰۹ perhour, and the duration of seizure prediction until seizure occurrence is ۳۰-۴۰ minutes.According to the results of the proposed method, the prediction of epileptic attacks has showna better performance compared to the compared related works, because it is possible to predictand prevent epileptic attacks in a longer time before the occurrence with higher accuracy andnew architecture

Keywords:

epilepsy , electroencephalogram(EEG) , wavelet transform(DWT) , intrinsic components(EMD) , support vector machine(SVM) , convolutional neural network(CCN).