Contextual Imputation With Missing Sequence of EEG Signals Using Generative Adversarial Networks

Missing values are very prevalent in real world; they are caused by various reasons such as user mistakes or device failures. They often cause critical problems especially in medical and healthcare application since they can lead to incorrect diagnosis or even cause system failure. Many of recent im...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Woonghee Lee, Jaeyoung Lee, Younghoon Kim
Formato: article
Lenguaje:EN
Publicado: IEEE 2021
Materias:
Acceso en línea:https://doaj.org/article/91c88ccd9a2f47b78b39c270b2c45d4b
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:91c88ccd9a2f47b78b39c270b2c45d4b
record_format dspace
spelling oai:doaj.org-article:91c88ccd9a2f47b78b39c270b2c45d4b2021-11-17T00:00:59ZContextual Imputation With Missing Sequence of EEG Signals Using Generative Adversarial Networks2169-353610.1109/ACCESS.2021.3126345https://doaj.org/article/91c88ccd9a2f47b78b39c270b2c45d4b2021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9606711/https://doaj.org/toc/2169-3536Missing values are very prevalent in real world; they are caused by various reasons such as user mistakes or device failures. They often cause critical problems especially in medical and healthcare application since they can lead to incorrect diagnosis or even cause system failure. Many of recent imputation techniques have adopted machine learning-based generative methods such as generative adversarial networks (GANs) to deal with missing values in medical data. They are, however, incapable of reproducing realistic time-series signals preserving important latent features such as sleep stages that are important context in many medical applications using electroencephalogram (EEG). In this study, we propose a novel GAN-based technique generating realistic EEG signal sequences which are not only shown natural but also correctly classified with sleep stages by implanting the latent features in the synthetic sequence. By experiments, we confirm that our model generates not only more realistic EEG signals than a recent GAN-based model but also preserve auxiliary information such as sleep stages. Furthermore, we demonstrate that existing machine learning methods based on EEG data still work well without sacrificing performance using the imputed data by using our method.Woonghee LeeJaeyoung LeeYounghoon KimIEEEarticleMissing data imputationelectroencephalogram (EEG)generated adversarial network (GAN)Electrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 151753-151765 (2021)
institution DOAJ
collection DOAJ
language EN
topic Missing data imputation
electroencephalogram (EEG)
generated adversarial network (GAN)
Electrical engineering. Electronics. Nuclear engineering
TK1-9971
spellingShingle Missing data imputation
electroencephalogram (EEG)
generated adversarial network (GAN)
Electrical engineering. Electronics. Nuclear engineering
TK1-9971
Woonghee Lee
Jaeyoung Lee
Younghoon Kim
Contextual Imputation With Missing Sequence of EEG Signals Using Generative Adversarial Networks
description Missing values are very prevalent in real world; they are caused by various reasons such as user mistakes or device failures. They often cause critical problems especially in medical and healthcare application since they can lead to incorrect diagnosis or even cause system failure. Many of recent imputation techniques have adopted machine learning-based generative methods such as generative adversarial networks (GANs) to deal with missing values in medical data. They are, however, incapable of reproducing realistic time-series signals preserving important latent features such as sleep stages that are important context in many medical applications using electroencephalogram (EEG). In this study, we propose a novel GAN-based technique generating realistic EEG signal sequences which are not only shown natural but also correctly classified with sleep stages by implanting the latent features in the synthetic sequence. By experiments, we confirm that our model generates not only more realistic EEG signals than a recent GAN-based model but also preserve auxiliary information such as sleep stages. Furthermore, we demonstrate that existing machine learning methods based on EEG data still work well without sacrificing performance using the imputed data by using our method.
format article
author Woonghee Lee
Jaeyoung Lee
Younghoon Kim
author_facet Woonghee Lee
Jaeyoung Lee
Younghoon Kim
author_sort Woonghee Lee
title Contextual Imputation With Missing Sequence of EEG Signals Using Generative Adversarial Networks
title_short Contextual Imputation With Missing Sequence of EEG Signals Using Generative Adversarial Networks
title_full Contextual Imputation With Missing Sequence of EEG Signals Using Generative Adversarial Networks
title_fullStr Contextual Imputation With Missing Sequence of EEG Signals Using Generative Adversarial Networks
title_full_unstemmed Contextual Imputation With Missing Sequence of EEG Signals Using Generative Adversarial Networks
title_sort contextual imputation with missing sequence of eeg signals using generative adversarial networks
publisher IEEE
publishDate 2021
url https://doaj.org/article/91c88ccd9a2f47b78b39c270b2c45d4b
work_keys_str_mv AT woongheelee contextualimputationwithmissingsequenceofeegsignalsusinggenerativeadversarialnetworks
AT jaeyounglee contextualimputationwithmissingsequenceofeegsignalsusinggenerativeadversarialnetworks
AT younghoonkim contextualimputationwithmissingsequenceofeegsignalsusinggenerativeadversarialnetworks
_version_ 1718426063088910336