TY - GEN
T1 - EEG-Based Seizure Detection Using Statistical and Spectral Features with Machine Learning
AU - Vakil, Amita Roshan
AU - Shetty, Mangala
AU - Shetty, Surendra
AU - Shetty, Spoorthi P.
AU - Pai, Shivanand
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - Information Technology (IT) has significantly influenced the healthcare sector in recent years, particularly in the field of neurological disorder diagnosis. EEG signal analysis, combined with machine learning and signal analysis, has become a valuable approach for the timely diagnosis and monitoring of epileptic seizures. EEG signals commonly deal with artifacts, low signal-to-noise ratio, and variability across patients, which pose challenges in accurately identifying seizure events. To break through these challenges, several automated approaches have been introduced to assist neurologists in accurately detecting seizures.This work presents an EEG-based seizure detection method using statistical and spectral features with a machine learning classifier. EEG data from the CHB-MIT dataset is preprocessed, segmented into 10-second epochs, and features are extracted. Visualization shows clear class separation. A Random Forest classifier achieves 97.67% accuracy, effectively identifying non-seizure events.
AB - Information Technology (IT) has significantly influenced the healthcare sector in recent years, particularly in the field of neurological disorder diagnosis. EEG signal analysis, combined with machine learning and signal analysis, has become a valuable approach for the timely diagnosis and monitoring of epileptic seizures. EEG signals commonly deal with artifacts, low signal-to-noise ratio, and variability across patients, which pose challenges in accurately identifying seizure events. To break through these challenges, several automated approaches have been introduced to assist neurologists in accurately detecting seizures.This work presents an EEG-based seizure detection method using statistical and spectral features with a machine learning classifier. EEG data from the CHB-MIT dataset is preprocessed, segmented into 10-second epochs, and features are extracted. Visualization shows clear class separation. A Random Forest classifier achieves 97.67% accuracy, effectively identifying non-seizure events.
UR - https://www.scopus.com/pages/publications/105030071978
UR - https://www.scopus.com/pages/publications/105030071978#tab=citedBy
U2 - 10.1109/DISCOVER66922.2025.11258978
DO - 10.1109/DISCOVER66922.2025.11258978
M3 - Conference contribution
AN - SCOPUS:105030071978
T3 - 2025 IEEE International Conference on Distributed Computing, VLSI, Electrical Circuits and Robotics, DISCOVER 2025 - Proceedings
SP - 277
EP - 282
BT - 2025 IEEE International Conference on Distributed Computing, VLSI, Electrical Circuits and Robotics, DISCOVER 2025 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 9th IEEE International Conference on Distributed Computing, VLSI, Electrical Circuits and Robotics, DISCOVER 2025
Y2 - 17 October 2025 through 18 October 2025
ER -