Epileptic Seizure Detection based on Different Events with XAI and Early Aid System for Patient Aid

The goal of this study is seizure detection in four class datasets for different seizure stages in epileptic patients. An early notification system is created to simulate the behavior of the patient experiencing a seizure receiving emergency assistance from caregivers, and a dataset acquired from Mendely is used to train different models. EEGNet’s effectiveness against hybrid deep learning models is demonstrated by the outcome. On the full dataset, Bi-GRU with attention, BI-Directional LSTM-GRU models, and conventional ensemble techniques like XGBoost can all do remarkably well. Channel 9 data is the most important feature, according to the SHAP interpretability analysis, which is conducted on several models with the aid of SHAP plots. The email SMTP system is used to simulate an early warning prototype. This method is essential for categorizing different seizure types according to occurrences in order to provide early warning and for developing a home automation strategy that will help victims.