表紙画像

Artifact removal from single-trial ERPs using non-Gaussian stochastic volatility models and particle filter

This paper considers improved modeling of artifactual noise for denoising of single-trial event-related potentials (ERPs) by state-space approach. Instead of the inadequate constant variance models used in existing studies, we propose to use stochastic volatility (SV) models to better describe the t...

詳細記述

保存先:
書誌詳細
主要な著者: Ting, Chee Ming, Shaikh Salleh, Sheikh Hussain, Zainuddin, Zaitul Marlizawati, Bahar, Arifah
フォーマット: 論文
出版事項: Institute of Electrical and Electronics Engineers Inc. 2014
主題:
QH Natural history
オンライン･アクセス:http://eprints.utm.my/id/eprint/51923/
http://dx.doi.org/10.1109/LSP.2014.2321000
タグ: タグ追加
タグなし, このレコードへの初めてのタグを付けませんか!
id my.utm.51923
record_format eprints
spelling my.utm.519232018-11-09T08:29:51Z http://eprints.utm.my/id/eprint/51923/ Artifact removal from single-trial ERPs using non-Gaussian stochastic volatility models and particle filter Ting, Chee Ming Shaikh Salleh, Sheikh Hussain Zainuddin, Zaitul Marlizawati Bahar, Arifah QH Natural history This paper considers improved modeling of artifactual noise for denoising of single-trial event-related potentials (ERPs) by state-space approach. Instead of the inadequate constant variance models used in existing studies, we propose to use stochastic volatility (SV) models to better describe the time-varying volatility in real ERP noise sources. We further propose a class of non-Gaussian SV models to capture the abrupt volatility changes typically present in impulsive noise, to improve artifact removal from ERPs. Two specifications are considered: (1) volatility driven by a heavy-tailed component and (2) transformation of volatility. Both result in volatility processes with heavy-tailed transition densities which can predict the impulsive noise volatility dynamics, more accurately than the Gaussian models. These SV noise models are incorporated in an autoregressive (AR) state-space ERP dynamic model. Parameter estimation is done using a Rao-Blackwellized particle filter (RBPF). Evaluation on simulated auditory brainstem responses (ABRs), corrupted by real eye-blink artifacts, shows that the non-Gaussian models can accurately detect the artifact-induced abrupt volatility spikes, and able to uncover the underlying inter-trial dynamics. Among them, the log-SV model performs the best. The results on real data demonstrate significant artifact suppression. Institute of Electrical and Electronics Engineers Inc. 2014 Article PeerReviewed Ting, Chee Ming and Shaikh Salleh, Sheikh Hussain and Zainuddin, Zaitul Marlizawati and Bahar, Arifah (2014) Artifact removal from single-trial ERPs using non-Gaussian stochastic volatility models and particle filter. IEEE Signal Processing Letters, 21 (8). pp. 923-927. ISSN 1070-9908 http://dx.doi.org/10.1109/LSP.2014.2321000
institution Universiti Teknologi Malaysia
building UTM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Teknologi Malaysia
content_source UTM Institutional Repository
url_provider http://eprints.utm.my/
topic QH Natural history
spellingShingle QH Natural history
Ting, Chee Ming
Shaikh Salleh, Sheikh Hussain
Zainuddin, Zaitul Marlizawati
Bahar, Arifah
Artifact removal from single-trial ERPs using non-Gaussian stochastic volatility models and particle filter
description This paper considers improved modeling of artifactual noise for denoising of single-trial event-related potentials (ERPs) by state-space approach. Instead of the inadequate constant variance models used in existing studies, we propose to use stochastic volatility (SV) models to better describe the time-varying volatility in real ERP noise sources. We further propose a class of non-Gaussian SV models to capture the abrupt volatility changes typically present in impulsive noise, to improve artifact removal from ERPs. Two specifications are considered: (1) volatility driven by a heavy-tailed component and (2) transformation of volatility. Both result in volatility processes with heavy-tailed transition densities which can predict the impulsive noise volatility dynamics, more accurately than the Gaussian models. These SV noise models are incorporated in an autoregressive (AR) state-space ERP dynamic model. Parameter estimation is done using a Rao-Blackwellized particle filter (RBPF). Evaluation on simulated auditory brainstem responses (ABRs), corrupted by real eye-blink artifacts, shows that the non-Gaussian models can accurately detect the artifact-induced abrupt volatility spikes, and able to uncover the underlying inter-trial dynamics. Among them, the log-SV model performs the best. The results on real data demonstrate significant artifact suppression.
format Article
author Ting, Chee Ming
Shaikh Salleh, Sheikh Hussain
Zainuddin, Zaitul Marlizawati
Bahar, Arifah
author_facet Ting, Chee Ming
Shaikh Salleh, Sheikh Hussain
Zainuddin, Zaitul Marlizawati
Bahar, Arifah
author_sort Ting, Chee Ming
title Artifact removal from single-trial ERPs using non-Gaussian stochastic volatility models and particle filter
title_short Artifact removal from single-trial ERPs using non-Gaussian stochastic volatility models and particle filter
title_full Artifact removal from single-trial ERPs using non-Gaussian stochastic volatility models and particle filter
title_fullStr Artifact removal from single-trial ERPs using non-Gaussian stochastic volatility models and particle filter
title_full_unstemmed Artifact removal from single-trial ERPs using non-Gaussian stochastic volatility models and particle filter
title_sort artifact removal from single-trial erps using non-gaussian stochastic volatility models and particle filter
publisher Institute of Electrical and Electronics Engineers Inc.
publishDate 2014
url http://eprints.utm.my/id/eprint/51923/
http://dx.doi.org/10.1109/LSP.2014.2321000
_version_ 1643653101680328704
score 13.251813

類似資料