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 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.

[1]  Viktor K. Jirsa,et al.  Cross-frequency coupling in real and virtual brain networks , 2013, Front. Comput. Neurosci..

[2]  Brian Litt,et al.  Epilepsyecosystem.org: crowd-sourcing reproducible seizure prediction with long-term human intracranial EEG , 2018, Brain : a journal of neurology.

[3]  Gerhard Nahler,et al.  first-in-man study , 2009 .

[4]  Robert C. Wolpert,et al.  A Review of the , 1985 .

[5]  E. Carrette,et al.  Functional brain connectivity from EEG in epilepsy: Seizure prediction and epileptogenic focus localization , 2014, Progress in Neurobiology.

[6]  F. Hayes,et al.  Lest we forget. , 1977, Military medicine.

[7]  S. Charpier,et al.  Slow modulations of high-frequency activity (40–140 Hz) discriminate preictal changes in human focal epilepsy , 2014, Scientific Reports.

[8]  Sridevi V. Sarma,et al.  Physiology of functional and effective networks in epilepsy , 2015, Clinical Neurophysiology.

[9]  Mohammed Yeasin,et al.  Learning Representations from EEG with Deep Recurrent-Convolutional Neural Networks , 2015, ICLR.

[10]  Nathalie Jette,et al.  Refractory epilepsy is a life-threatening disease , 2016, Neurology.

[11]  William C. Stacey,et al.  Seizure Prediction Is Possible–Now Let's Make It Practical , 2018, EBioMedicine.

[12]  Mohamad Sawan,et al.  A Functional-Genetic Scheme for Seizure Forecasting in Canine Epilepsy , 2018, IEEE Transactions on Biomedical Engineering.

[13]  K. Lehnertz,et al.  Seizure prediction — ready for a new era , 2018, Nature Reviews Neurology.

[14]  Sandy Rihana,et al.  Bilateral preictal signature of phase-amplitude coupling in canine epilepsy , 2018, Epilepsy Research.

[15]  F. Mormann,et al.  Seizure prediction for therapeutic devices: A review , 2016, Journal of Neuroscience Methods.

[16]  Brian Litt,et al.  Forecasting Seizures in Dogs with Naturally Occurring Epilepsy , 2014, PloS one.

[17]  Mohamad Sawan,et al.  Towards accurate prediction of epileptic seizures: A review , 2017, Biomed. Signal Process. Control..

[18]  Sabato Santaniello,et al.  Decision Support System for Seizure Onset Zone Localization Based on Channel Ranking and High-Frequency EEG Activity , 2019, IEEE Journal of Biomedical and Health Informatics.

[19]  Michalis E. Zervakis,et al.  A Long Short-Term Memory deep learning network for the prediction of epileptic seizures using EEG signals , 2018, Comput. Biol. Medicine.

[20]  Benjamin H. Brinkmann,et al.  Towards Improved Design and Evaluation of Epileptic Seizure Predictors , 2018, IEEE Transactions on Biomedical Engineering.

[21]  B. Litt,et al.  A novel implanted device to wirelessly record and analyze continuous intracranial canine EEG , 2011, Epilepsy Research.

[22]  Rabab K. Ward,et al.  Optimized deep neural network architecture for robust detection of epileptic seizures using EEG signals , 2019, Clinical Neurophysiology.

[23]  Matthew J. Hausknecht,et al.  Beyond short snippets: Deep networks for video classification , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[24]  C. Hancock.  The long and winding road , 2005, Nature Reviews Drug Discovery.

[25]  Philip Grewe,et al.  Trends in epilepsy surgery: stable surgical numbers despite increasing presurgical volumes , 2016, Journal of Neurology, Neurosurgery & Psychiatry.

[26]  F. Mormann,et al.  Seizure prediction: the long and winding road. , 2007, Brain : a journal of neurology.

[27]  Josemir W Sander,et al.  Comorbidities of epilepsy: current concepts and future perspectives , 2016, The Lancet Neurology.

[28]  N. Crone,et al.  Network dynamics of the brain and influence of the epileptic seizure onset zone , 2014, Proceedings of the National Academy of Sciences.

[29]  Dean R. Freestone,et al.  A forward-looking review of seizure prediction , 2017, Current opinion in neurology.

[30]  Joelle Pineau,et al.  Learning Robust Features using Deep Learning for Automatic Seizure Detection , 2016, MLHC.

[31]  R. Gross.  Spotlight on the January 5 Issue , 2016, Neurology.

[32]  Joost B. Wagenaar,et al.  Forecasting Seizures Using Intracranial EEG Measures and SVM in Naturally Occurring Canine Epilepsy , 2015, PloS one.

[33]  David M. Himes,et al.  Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: a first-in-man study , 2013, The Lancet Neurology.

[34]  Jedediah M. Singer,et al.  Quickest detection of drug-resistant seizures: An optimal control approach , 2011, Epilepsy & Behavior.

[35]  Sandy Rihana,et al.  Bispectrum Features and Multilayer Perceptron Classifier to Enhance Seizure Prediction , 2018, Scientific Reports.