Epileptic Seizure Prediction using Deep Learning

Mr. L. Kiran Kumar
Assistant Professor.Department of Information Technology Department
Maharaj Vijayaram Gajapathi Raj
College of Engineering(A) Chintalavalasa, Vizianagaram, Andhra Pradesh, India.
kirankumar@mvgrce.edu.in
Besetty Shubham
Department of Information Technology Maharaj Vijayaram Gajapathi Raj college of Engineering(A) Chintalavalasa, Vizianagaram, Andhra
Pradesh, India.shubhambesetty@gmail.com
Jagarapu Joshna
Department of Information Technology Maharaj Vijayaram Gajapathi Raj college of Engineering(A) Chintalavalasa, Vizianagaram,Andhra Pradesh, India.testmailjoshna@gmail.com
Madisa Abhilash
Department ofInformation Technology  Maharaj Vijayaram Gajapathi Raj college of Engineering(A) Chintalavalasa, Vizianagaram,Andhra Pradesh, India.abhilashmadisa@gmail.com
Gudivada Avinash
Department of Information Technology Maharaj Vijayaram Gajapathi Raj college of Engineering(A) Chintalavalasa, Vizianagaram, Andhra
Pradesh, India.22331a1233@gmail.com

Abstract— Epilepsy is a neurological disorder characterized by sudden and recurrent seizures caused by abnormal electrical activity in the brain. Early prediction of seizures is crucial to improve patient safety and enable timely medical intervention. Traditional seizure detectionmethods rely on manual analysis of Electroencephalogram (EEG) signals, which is time-consuming and prone to human error. This paper proposes a deep learning-based epileptic seizure prediction system using Long Short-Term Memory (LSTM) networks. The system processes EEG data through preprocessing techniques such as noise removal, normalization, and feature extraction. The LSTM model learns temporal patterns in EEG signals and predicts seizure occurrence with high accuracy.The model achieved an accuracy of 98.53%, precision of 0.9837, recall of0.9873, and ROC-AUC score of 0.9967, demonstrating strong predictive performance. A user-friendly interface is  developed using Streamlit, allowing real-time prediction byuploading EEG data. This system provides an efficient,accurate, and accessible solution for seizure prediction andmonitoring.
Keywords— Deep Learning, EEG, Epileptic Seizure,LSTM, Prediction, Streamlit.