Cargando…

Obstructive Sleep Apnea Recognition Based on Multi-Bands Spectral Entropy Analysis of Short-Time Heart Rate Variability

Obstructive sleep apnea (OSA) syndrome is a common sleep disorder. As an alternative to polysomnography (PSG) for OSA screening, the current automatic OSA detection methods mainly concentrate on feature extraction and classifier selection based on physiological signals. It has been reported that OSA...

Descripción completa

Detalles Bibliográficos
Autores principales: Shao, Shiliang, Wang, Ting, Song, Chunhe, Chen, Xingchi, Cui, Enuo, Zhao, Hai
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7515341/
https://www.ncbi.nlm.nih.gov/pubmed/33267526
http://dx.doi.org/10.3390/e21080812
_version_ 1783586795132289024
author Shao, Shiliang
Wang, Ting
Song, Chunhe
Chen, Xingchi
Cui, Enuo
Zhao, Hai
author_facet Shao, Shiliang
Wang, Ting
Song, Chunhe
Chen, Xingchi
Cui, Enuo
Zhao, Hai
author_sort Shao, Shiliang
collection PubMed
description Obstructive sleep apnea (OSA) syndrome is a common sleep disorder. As an alternative to polysomnography (PSG) for OSA screening, the current automatic OSA detection methods mainly concentrate on feature extraction and classifier selection based on physiological signals. It has been reported that OSA is, along with autonomic nervous system (ANS) dysfunction and heart rate variability (HRV), a useful tool for ANS assessment. Therefore, in this paper, eight novel indices of short-time HRV are extracted for OSA detection, which are based on the proposed multi-bands time-frequency spectrum entropy (MTFSE) method. In the MTFSE, firstly, the power spectrum of HRV is estimated by the Burg–AR model, and the time-frequency spectrum image (TFSI) is obtained. Secondly, according to the physiological significance of HRV, the TFSI is divided into multiple sub-bands according to frequency. Last but not least, by studying the Shannon entropy of different sub-bands and the relationships among them, the eight indices are obtained. In order to validate the performance of MTFSE-based indices, the Physionet Apnea–ECG database and K-nearest neighbor (KNN), support vector machine (SVM), and decision tree (DT) classification methods are used. The SVM classification method gets the highest classification accuracy, its average accuracy is 91.89%, the average sensitivity is 88.01%, and the average specificity is 93.98%. Undeniably, the MTFSE-based indices provide a novel idea for the screening of OSA disease.
format Online
Article
Text
id pubmed-7515341
institution National Center for Biotechnology Information
language English
publishDate 2019
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-75153412020-11-09 Obstructive Sleep Apnea Recognition Based on Multi-Bands Spectral Entropy Analysis of Short-Time Heart Rate Variability Shao, Shiliang Wang, Ting Song, Chunhe Chen, Xingchi Cui, Enuo Zhao, Hai Entropy (Basel) Article Obstructive sleep apnea (OSA) syndrome is a common sleep disorder. As an alternative to polysomnography (PSG) for OSA screening, the current automatic OSA detection methods mainly concentrate on feature extraction and classifier selection based on physiological signals. It has been reported that OSA is, along with autonomic nervous system (ANS) dysfunction and heart rate variability (HRV), a useful tool for ANS assessment. Therefore, in this paper, eight novel indices of short-time HRV are extracted for OSA detection, which are based on the proposed multi-bands time-frequency spectrum entropy (MTFSE) method. In the MTFSE, firstly, the power spectrum of HRV is estimated by the Burg–AR model, and the time-frequency spectrum image (TFSI) is obtained. Secondly, according to the physiological significance of HRV, the TFSI is divided into multiple sub-bands according to frequency. Last but not least, by studying the Shannon entropy of different sub-bands and the relationships among them, the eight indices are obtained. In order to validate the performance of MTFSE-based indices, the Physionet Apnea–ECG database and K-nearest neighbor (KNN), support vector machine (SVM), and decision tree (DT) classification methods are used. The SVM classification method gets the highest classification accuracy, its average accuracy is 91.89%, the average sensitivity is 88.01%, and the average specificity is 93.98%. Undeniably, the MTFSE-based indices provide a novel idea for the screening of OSA disease. MDPI 2019-08-20 /pmc/articles/PMC7515341/ /pubmed/33267526 http://dx.doi.org/10.3390/e21080812 Text en © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Shao, Shiliang
Wang, Ting
Song, Chunhe
Chen, Xingchi
Cui, Enuo
Zhao, Hai
Obstructive Sleep Apnea Recognition Based on Multi-Bands Spectral Entropy Analysis of Short-Time Heart Rate Variability
title Obstructive Sleep Apnea Recognition Based on Multi-Bands Spectral Entropy Analysis of Short-Time Heart Rate Variability
title_full Obstructive Sleep Apnea Recognition Based on Multi-Bands Spectral Entropy Analysis of Short-Time Heart Rate Variability
title_fullStr Obstructive Sleep Apnea Recognition Based on Multi-Bands Spectral Entropy Analysis of Short-Time Heart Rate Variability
title_full_unstemmed Obstructive Sleep Apnea Recognition Based on Multi-Bands Spectral Entropy Analysis of Short-Time Heart Rate Variability
title_short Obstructive Sleep Apnea Recognition Based on Multi-Bands Spectral Entropy Analysis of Short-Time Heart Rate Variability
title_sort obstructive sleep apnea recognition based on multi-bands spectral entropy analysis of short-time heart rate variability
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7515341/
https://www.ncbi.nlm.nih.gov/pubmed/33267526
http://dx.doi.org/10.3390/e21080812
work_keys_str_mv AT shaoshiliang obstructivesleepapnearecognitionbasedonmultibandsspectralentropyanalysisofshorttimeheartratevariability
AT wangting obstructivesleepapnearecognitionbasedonmultibandsspectralentropyanalysisofshorttimeheartratevariability
AT songchunhe obstructivesleepapnearecognitionbasedonmultibandsspectralentropyanalysisofshorttimeheartratevariability
AT chenxingchi obstructivesleepapnearecognitionbasedonmultibandsspectralentropyanalysisofshorttimeheartratevariability
AT cuienuo obstructivesleepapnearecognitionbasedonmultibandsspectralentropyanalysisofshorttimeheartratevariability
AT zhaohai obstructivesleepapnearecognitionbasedonmultibandsspectralentropyanalysisofshorttimeheartratevariability