Automatic Detection of Atrial Fibrillation from Ballistocardiogram (BCG) Using Wavelet Features and Machine Learning

Bin Yu; Biyong Zhang; Lisheng Xu; Peng Fang; Jun Hu

doi:10.1109/EMBC.2019.8857059

Automatic Detection of Atrial Fibrillation from Ballistocardiogram (BCG) Using Wavelet Features and Machine Learning

Bin Yu, Biyong Zhang, Lisheng Xu, Peng Fang, Jun Hu

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

14 Citaten (Scopus)

Samenvatting

This paper presents an unobtrusive method for automatic detection of atrial fibrillation (AF) from single-channel ballistocardiogram (BCG) recordings during sleep. We developed a remote data acquisition system that measures BCG signals through an electromechanical-film sensor embedded into a bed's mattress and transmits the BCG data to a remote database on the cloud server. In the feasibility study, 12 AF patients' data were recorded during entire night of sleep. Each BCG recording was split into nonoverlapping 30s epochs labeled either AF or normal. Using the features extracted from stationary wavelet transform of these epochs, three popular machine learning classifiers (support vector machine, K-nearest neighbor, and ensembles) have been trained and evaluated on the set of 7816 epochs employing 30% hold-out validation. The results showed that all the trained classifiers could achieve an accuracy rate above 91.5%. The optimized ensembles model (Bagged Trees) could achieve accuracy, sensitivity, and specificity of 0.944, 0.970 and 0.891, respectively. These results suggest that the proposed BCG-based AF detection can be a potential initial screening and detection tool of AF in home-monitoring applications.

Originele taal-2	Engels
Titel	2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019
Uitgeverij	Institute of Electrical and Electronics Engineers
Pagina's	4322-4325
Aantal pagina's	4
ISBN van elektronische versie	9781538613115
DOI's	https://doi.org/10.1109/EMBC.2019.8857059
Status	Gepubliceerd - jul. 2019
Evenement	41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019 - City Cube Berlin, Berlin, Duitsland Duur: 23 jul. 2019 → 27 jul. 2019 https://embc.embs.org/2019/

Congres

Congres	41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019
Verkorte titel	EMBC 2019
Land/Regio	Duitsland
Stad	Berlin
Periode	23/07/19 → 27/07/19
Internet adres	https://embc.embs.org/2019/

Financiering

1008 10.1109/EMBC.2019.8857059 0b0000648b68ffcd Active orig-research F T F F X F T F Publish IEEE This paper presents an unobtrusive method for automatic detection of atrial fibrillation (AF) from single-channel ballistocardiogram (BCG) recordings during sleep. We developed a remote data acquisition system that measures BCG signals through an electromechanical-film sensor embedded into a bed’s mattress and transmits the BCG data to a remote database on the cloud server. In the feasibility study, 12 AF patients’ data were recorded during entire night of sleep. Each BCG recording was split into nonoverlapping 30s epochs labeled either AF or normal. Using the features extracted from stationary wavelet transform of these epochs, three popular machine learning classifiers (support vector machine, K-nearest neighbor, and ensembles) have been trained and evaluated on the set of 7816 epochs employing 30% hold-out validation. The results showed that all the trained classifiers could achieve an accuracy rate above 91.5%. The optimized ensembles model (Bagged Trees) could achieve accuracy, sensitivity, and specificity of 0.944, 0.970 and 0.891, respectively. These results suggest that the proposed BCG-based AF detection can be a potential initial screening and detection tool of AF in home-monitoring applications. 1 Yu, Bin Bin Yu Bin Yu Industrial Design, TU/e, 5612 AZ, the Netherlands Author 2 Zhang, Biyong Biyong Zhang Biyong Zhang Industrial Design, TU/e, 5612 AZ, the Netherlands Author 3 Xu, Lisheng Lisheng Xu Lisheng Xu Sino-Dutch Biomedical and Information Engineering, Northeastern University, Shenyang, 110169, China Author 4 Fang, Peng Peng Fang Peng Fang Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China Author 5 Hu, Jun Jun Hu Jun Hu Industrial Design, TU/e, 5612 AZ, the Netherlands Author 2019 July 2019 10 2 840703 08857059.pdf 4322-4325 8857059 Feature extraction Electrocardiography Mathematical model Atrial fibrillation Support vector machines Monitoring Brain modeling Atrial fibrillation (AF) Ballistocardiogram (BCG) Machine Learning

Toegang tot document

10.1109/EMBC.2019.8857059

Andere bestanden en links

Link to publication in Scopus

Sleep Medicine
van Gilst, M. M. (Content manager) & van der Hout-van der Jagt, M. B. (Content manager)
Impact: Research Topic/Theme (at group level)

Citeer dit

Yu, B., Zhang, B., Xu, L., Fang, P., & Hu, J. (2019). Automatic Detection of Atrial Fibrillation from Ballistocardiogram (BCG) Using Wavelet Features and Machine Learning. In 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019 (blz. 4322-4325). Artikel 8857059 Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/EMBC.2019.8857059

@inproceedings{667a11a296074db2b0ae320f6400f7dc,

title = "Automatic Detection of Atrial Fibrillation from Ballistocardiogram (BCG) Using Wavelet Features and Machine Learning",

abstract = "This paper presents an unobtrusive method for automatic detection of atrial fibrillation (AF) from single-channel ballistocardiogram (BCG) recordings during sleep. We developed a remote data acquisition system that measures BCG signals through an electromechanical-film sensor embedded into a bed's mattress and transmits the BCG data to a remote database on the cloud server. In the feasibility study, 12 AF patients' data were recorded during entire night of sleep. Each BCG recording was split into nonoverlapping 30s epochs labeled either AF or normal. Using the features extracted from stationary wavelet transform of these epochs, three popular machine learning classifiers (support vector machine, K-nearest neighbor, and ensembles) have been trained and evaluated on the set of 7816 epochs employing 30% hold-out validation. The results showed that all the trained classifiers could achieve an accuracy rate above 91.5%. The optimized ensembles model (Bagged Trees) could achieve accuracy, sensitivity, and specificity of 0.944, 0.970 and 0.891, respectively. These results suggest that the proposed BCG-based AF detection can be a potential initial screening and detection tool of AF in home-monitoring applications.",

keywords = "Atrial fibrillation (AF), Ballistocardiogram (BCG), Machine Learning, Algorithms, Sleep, Wavelet Analysis, Humans, Ballistocardiography, Atrial Fibrillation/diagnosis, Monitoring, Physiologic/methods",

author = "Bin Yu and Biyong Zhang and Lisheng Xu and Peng Fang and Jun Hu",

year = "2019",

month = jul,

doi = "10.1109/EMBC.2019.8857059",

language = "English",

pages = "4322--4325",

booktitle = "2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019<br/>, EMBC 2019 ; Conference date: 23-07-2019 Through 27-07-2019",

url = "https://embc.embs.org/2019/",

}

Yu, B, Zhang, B, Xu, L, Fang, P & Hu, J 2019, Automatic Detection of Atrial Fibrillation from Ballistocardiogram (BCG) Using Wavelet Features and Machine Learning. in 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019., 8857059, Institute of Electrical and Electronics Engineers, blz. 4322-4325, 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019
, Berlin, Berlijn, Duitsland, 23/07/19. https://doi.org/10.1109/EMBC.2019.8857059

Automatic Detection of Atrial Fibrillation from Ballistocardiogram (BCG) Using Wavelet Features and Machine Learning. / Yu, Bin; Zhang, Biyong; Xu, Lisheng et al.
2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019. Institute of Electrical and Electronics Engineers, 2019. blz. 4322-4325 8857059.

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

TY - GEN

T1 - Automatic Detection of Atrial Fibrillation from Ballistocardiogram (BCG) Using Wavelet Features and Machine Learning

AU - Yu, Bin

AU - Zhang, Biyong

AU - Xu, Lisheng

AU - Fang, Peng

AU - Hu, Jun

PY - 2019/7

Y1 - 2019/7

N2 - This paper presents an unobtrusive method for automatic detection of atrial fibrillation (AF) from single-channel ballistocardiogram (BCG) recordings during sleep. We developed a remote data acquisition system that measures BCG signals through an electromechanical-film sensor embedded into a bed's mattress and transmits the BCG data to a remote database on the cloud server. In the feasibility study, 12 AF patients' data were recorded during entire night of sleep. Each BCG recording was split into nonoverlapping 30s epochs labeled either AF or normal. Using the features extracted from stationary wavelet transform of these epochs, three popular machine learning classifiers (support vector machine, K-nearest neighbor, and ensembles) have been trained and evaluated on the set of 7816 epochs employing 30% hold-out validation. The results showed that all the trained classifiers could achieve an accuracy rate above 91.5%. The optimized ensembles model (Bagged Trees) could achieve accuracy, sensitivity, and specificity of 0.944, 0.970 and 0.891, respectively. These results suggest that the proposed BCG-based AF detection can be a potential initial screening and detection tool of AF in home-monitoring applications.

AB - This paper presents an unobtrusive method for automatic detection of atrial fibrillation (AF) from single-channel ballistocardiogram (BCG) recordings during sleep. We developed a remote data acquisition system that measures BCG signals through an electromechanical-film sensor embedded into a bed's mattress and transmits the BCG data to a remote database on the cloud server. In the feasibility study, 12 AF patients' data were recorded during entire night of sleep. Each BCG recording was split into nonoverlapping 30s epochs labeled either AF or normal. Using the features extracted from stationary wavelet transform of these epochs, three popular machine learning classifiers (support vector machine, K-nearest neighbor, and ensembles) have been trained and evaluated on the set of 7816 epochs employing 30% hold-out validation. The results showed that all the trained classifiers could achieve an accuracy rate above 91.5%. The optimized ensembles model (Bagged Trees) could achieve accuracy, sensitivity, and specificity of 0.944, 0.970 and 0.891, respectively. These results suggest that the proposed BCG-based AF detection can be a potential initial screening and detection tool of AF in home-monitoring applications.

KW - Atrial fibrillation (AF)

KW - Ballistocardiogram (BCG)

KW - Machine Learning

KW - Algorithms

KW - Sleep

KW - Wavelet Analysis

KW - Humans

KW - Ballistocardiography

KW - Atrial Fibrillation/diagnosis

KW - Monitoring, Physiologic/methods

UR - http://www.scopus.com/inward/record.url?scp=85077901035&partnerID=8YFLogxK

U2 - 10.1109/EMBC.2019.8857059

DO - 10.1109/EMBC.2019.8857059

M3 - Conference contribution

C2 - 31946824

AN - SCOPUS:85077901035

SP - 4322

EP - 4325

BT - 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019

PB - Institute of Electrical and Electronics Engineers

T2 - 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019<br/>

Y2 - 23 July 2019 through 27 July 2019

ER -

Yu B, Zhang B, Xu L, Fang P, Hu J. Automatic Detection of Atrial Fibrillation from Ballistocardiogram (BCG) Using Wavelet Features and Machine Learning. In 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019. Institute of Electrical and Electronics Engineers. 2019. blz. 4322-4325. 8857059 doi: 10.1109/EMBC.2019.8857059

Automatic Detection of Atrial Fibrillation from Ballistocardiogram (BCG) Using Wavelet Features and Machine Learning

Samenvatting

Congres

Financiering

Toegang tot document

Andere bestanden en links

Vingerafdruk

Impact

Sleep Medicine

Citeer dit