Abstract
Epilepsy is one of the most common neurological brain dysfunctions after stroke, Alzheimer and migraine in humans. Epilepsy is a disease that affects approximately 50 million people worldwide. Seizure is occurred due to abnormal electrical discharges of the neurons in the brain cell. EEG is commonly used to accommodate information about the electrical activity of the brain, an automated epilepsy seizure detection and classification using EEG signals in deep learning techniques. As the rapid innovation in the field of healthcare increases, various biomedical signals, namely electrocardiogram, electroencephalogram (EEG) and electromyogram, play a crucial role for the measurement of various diseases such as cardiovascular diseases, brain disorders, etc. We have proposed a novel deep convolutional neural network model, which classifies the EEG signals into two distinct classes namely normal and seizure. The experiment is performed on a publicly available benchmark CHB-MIT database. The performance of the model was evaluated in terms of sensitivity, specificity and accuracy. The experimental results that have been obtained are 87.40% accuracy, 88.10% sensitivity and 87.10% specificity, which are then compared to other existing works of literature and are better than the existing work.
