Bispectrum and Recurrent Neural Networks: Improved Classification of Interictal and Preictal States

This work proposes a novel approach for the classification of interictal and preictal brain states based on bispectrum analysis and recurrent Long Short-Term Memory (LSTM) neural networks. Two features were first extracted from bilateral intracranial electroencephalography (iEEG) recordings of dogs...

Full description

Saved in:
Bibliographic Details
Published in:Scientific reports 2019-10, Vol.9 (1), p.15649-9, Article 15649
Main Authors: Gagliano, Laura, Bou Assi, Elie, Nguyen, Dang K., Sawan, Mohamad
Format: Article
Language:English
Subjects:
639/166/985
692/617/375/178
Animals
Classification
Comparative analysis
Dogs
EEG
Electroencephalography
Epilepsy
Epilepsy - pathology
Epilepsy - physiopathology
Humanities and Social Sciences
Humans
Long short-term memory
Memory, Short-Term
Models, Neurological
multidisciplinary
Nerve Net - pathology
Nerve Net - physiopathology
Neural networks
Science
Science (multidisciplinary)
Seizures
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This work proposes a novel approach for the classification of interictal and preictal brain states based on bispectrum analysis and recurrent Long Short-Term Memory (LSTM) neural networks. Two features were first extracted from bilateral intracranial electroencephalography (iEEG) recordings of dogs with naturally occurring focal epilepsy. Single-layer LSTM networks were trained to classify 5-min long feature vectors as preictal or interictal. Classification performances were compared to previous work involving multilayer perceptron networks and higher-order spectral (HOS) features on the same dataset. The proposed LSTM network proved superior to the multilayer perceptron network and achieved an average classification accuracy of 86.29% on held-out data. Results imply the possibility of forecasting epileptic seizures using recurrent neural networks, with minimal feature extraction.
ISSN:2045-2322
2045-2322
DOI:10.1038/s41598-019-52152-2