Epileptic Seizure Prediction With Multi-View Convolutional Neural Networks

The unpredictability of seizures is often considered by patients to be the most problematic aspect of epilepsy, so this work aims to develop an accurate epilepsy seizure predictor, making it possible to enable devices to warn patients of impeding seizures. To develop a model for seizure prediction,...

Full description

Saved in:
Bibliographic Details
Main Authors: Chien-Liang Liu, Bin Xiao, Wen-Hoar Hsaio, Vincent S. Tseng
Format: Article
Language:English
Published: IEEE 2019-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/8910555/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849738639285682176
author Chien-Liang Liu
Bin Xiao
Wen-Hoar Hsaio
Vincent S. Tseng
author_facet Chien-Liang Liu
Bin Xiao
Wen-Hoar Hsaio
Vincent S. Tseng
author_sort Chien-Liang Liu
collection DOAJ
description The unpredictability of seizures is often considered by patients to be the most problematic aspect of epilepsy, so this work aims to develop an accurate epilepsy seizure predictor, making it possible to enable devices to warn patients of impeding seizures. To develop a model for seizure prediction, most studies relied on Electroencephalograms (EEGs) to capture physiological measurements of epilepsy. This work uses the two domains of EEGs, including frequency domain and time domain, to provide two different views for the same data source. Subsequently, this work proposes a multi-view convolutional neural network framework to predict the occurrence of epilepsy seizures with the goal of acquiring a shared representation of time-domain and frequency-domain features. By conducting experiments on Kaggle data set, we demonstrated that the proposed method outperforms all methods listed in the Kaggle leader board. Additionally, our proposed model achieves average area under the curve (AUCs) of 0.82 and 0.89 on two subjects of CHB-MIT scalp EEG data set. This work serves as an effective paradigm for applying deep learning approaches to the crucial topic of risk prediction in health domains.
format Article
id doaj-art-97ff43fe4e1f490c87417cda4e4b6ceb
institution DOAJ
issn 2169-3536
language English
publishDate 2019-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-97ff43fe4e1f490c87417cda4e4b6ceb2025-08-20T03:06:30ZengIEEEIEEE Access2169-35362019-01-01717035217036110.1109/ACCESS.2019.29552858910555Epileptic Seizure Prediction With Multi-View Convolutional Neural NetworksChien-Liang Liu0https://orcid.org/0000-0002-2724-7199Bin Xiao1https://orcid.org/0000-0003-1992-4214Wen-Hoar Hsaio2https://orcid.org/0000-0002-4439-676XVincent S. Tseng3https://orcid.org/0000-0001-5449-7700Department of Industrial Engineering and Management, National Chiao Tung University, Hsinchu, TaiwanDepartment of Computer Science, National Chiao Tung University, Hsinchu, TaiwanInformation Management Center, National Chung-Shan Institute of Science and Technology, Taoyuan City, TaiwanDepartment of Computer Science, National Chiao Tung University, Hsinchu, TaiwanThe unpredictability of seizures is often considered by patients to be the most problematic aspect of epilepsy, so this work aims to develop an accurate epilepsy seizure predictor, making it possible to enable devices to warn patients of impeding seizures. To develop a model for seizure prediction, most studies relied on Electroencephalograms (EEGs) to capture physiological measurements of epilepsy. This work uses the two domains of EEGs, including frequency domain and time domain, to provide two different views for the same data source. Subsequently, this work proposes a multi-view convolutional neural network framework to predict the occurrence of epilepsy seizures with the goal of acquiring a shared representation of time-domain and frequency-domain features. By conducting experiments on Kaggle data set, we demonstrated that the proposed method outperforms all methods listed in the Kaggle leader board. Additionally, our proposed model achieves average area under the curve (AUCs) of 0.82 and 0.89 on two subjects of CHB-MIT scalp EEG data set. This work serves as an effective paradigm for applying deep learning approaches to the crucial topic of risk prediction in health domains.https://ieeexplore.ieee.org/document/8910555/Electroencephalograms (EEG)seizure predictionconvolutional neural network (CNN)multi-view CNNrepresentation learning
spellingShingle Chien-Liang Liu
Bin Xiao
Wen-Hoar Hsaio
Vincent S. Tseng
Epileptic Seizure Prediction With Multi-View Convolutional Neural Networks
IEEE Access
Electroencephalograms (EEG)
seizure prediction
convolutional neural network (CNN)
multi-view CNN
representation learning
title Epileptic Seizure Prediction With Multi-View Convolutional Neural Networks
title_full Epileptic Seizure Prediction With Multi-View Convolutional Neural Networks
title_fullStr Epileptic Seizure Prediction With Multi-View Convolutional Neural Networks
title_full_unstemmed Epileptic Seizure Prediction With Multi-View Convolutional Neural Networks
title_short Epileptic Seizure Prediction With Multi-View Convolutional Neural Networks
title_sort epileptic seizure prediction with multi view convolutional neural networks
topic Electroencephalograms (EEG)
seizure prediction
convolutional neural network (CNN)
multi-view CNN
representation learning
url https://ieeexplore.ieee.org/document/8910555/
work_keys_str_mv AT chienliangliu epilepticseizurepredictionwithmultiviewconvolutionalneuralnetworks
AT binxiao epilepticseizurepredictionwithmultiviewconvolutionalneuralnetworks
AT wenhoarhsaio epilepticseizurepredictionwithmultiviewconvolutionalneuralnetworks
AT vincentstseng epilepticseizurepredictionwithmultiviewconvolutionalneuralnetworks