Extremely random forest based automatic tonic-clonic seizure detection using spectral analysis on electroencephalography data.

Stewart, Craig; Fung, Wai Keung; Fough, Nazila; Prabhu, Radhakrishna

Abstract

Machine learning proliferates society and has begun changing medicine. This report covers an investigation into how Extremely Random Forests combined with Fast Fourier Transform feature extraction performed on two-dimensional time-series Epileptic Seizure data from the Bonn/UCI dataset. It found that robust classification can take place with lower channel counts, achieving 99.81% recall, 98.8% precision and 99.35% accuracy, outperforming previous works carried into this scenario.

Citation

STEWART, C., FUNG, WAI KEUNG, FOUGH, N. and PRABHU, R. 2023. Extremely random forest based automatic tonic-clonic seizure detection using spectral analysis on electroencephalography data. In Proceedings of the 21st IEEE (Institute of Electrical and Electronics Engineers) Interregional NEWCAS conference 2023 (NEWCAS 2023), 26-28 June 2023, Edinburgh, UK. Piscataway: IEEE [online], article 10198101. Available from: https://doi.org/10.1109/NEWCAS57931.2023.10198101

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 21st IEEE (Institute of Electrical and Electronics Engineers) Interregional NEWCAS conference 2023 (NEWCAS 2023)
• Start Date: Jun 26, 2023
• End Date: Jun 28, 2023
• Acceptance Date: Apr 12, 2023
• Online Publication Date: Aug 7, 2023
• Publication Date: Dec 31, 2023
• Deposit Date: Sep 4, 2023
• Publicly Available Date: Sep 4, 2023
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Peer Reviewed: Peer Reviewed
• Series ISSN: 2474-9672
• ISBN: 9798350300246
• DOI: https://doi.org/10.1109/NEWCAS57931.2023.10198101
• Keywords: Epilepsy; Extremely random forest; Electroencephalography; Fourier transform
• Public URL: https://rgu-repository.worktribe.com/output/2054236