An Automated Epilepsy Seizure Detection System (AESD) Using Deep Learning Models

Epilepsy is a brain disorder that results in abnormal electrical activity in the brain. Epilepsy is also called a seizure disorder. The diagnosis of epilepsy can be made if the person suffers from two or more seizures. Seizures are the main symptom of epilepsy. Different models are introduced to diagnose epileptic seizures with the help of electroencephalography (EEG). EEG is wavelet signals that are most widely used to find the abnormal events identified in the brain. Deep Learning (DL) algorithms can efficiently discover epileptic seizures based on their features. In this study, an automated epilepsy seizure detection system (AESD) is developed to diagnose and detect epilepsy seizures from the human brain. A pre-trained DL model VGG-19 is used to train the EEG sample datasets. VGG-19 helps to extract the features of ECG samples and transfer the data to the proposed approach for better outcomes. The existing models analyze many challenges to seeing this disease. The dataset attributes include patient family history, age, and information about medicines that have been used for a long time. The experiments use Python libraries for DL algorithms, such as Keras, Pandas, Matplotlib, and Numpy. The performance of the proposed approach is analyzed based on parameters such as accuracy, sensitivity, specificity, and F1-score.