Seizure Recognition using SVM

Seizure Recognition using SVM

Mitali Wadhwani1, Hetvi Patel2 , Nandini Chaudhary 3

123Research Scholar, Institute of Information Technology, Krishna School Of Emerging Technology & Applied Research, KPGU University , Varnama , Vadodara, Gujarat, India

4 Assistant Professor, Institute of Information Technology, Krishna School Of Emerging Technology & Applied Research, KPGU University , Varnama , Vadodara, Gujarat, India

Abstract -  This study focuses on using machine learning, specifically Support Vector Machine (SVM) models, for automatic recognition of epileptic seizures from EEG signals. Preprocessed EEG data with relevant features was used to train and evaluate the SVM classifier, achieving promising results in distinguishing between seizure and non-seizure events. The model's effectiveness was demonstrated through performance metrics like accuracy and confusion matrices. This study highlights the potential for traditional machine learning methods in real-time seizure detection, laying the groundwork for future enhancements such as wearable deployment, improved predictive models, and early seizure forecasting.

Key Words: Support Vector Machine (SVM), epileptic seizures, EEG signals, preprocessed EEG data, distinguish seizure and non-seizure events, performance metrics, accuracy, confusion matrices, real-time seizure detection.