Model-based data-driven approach for sleep apnea detection

Sandeep Gutta, Qi Cheng, Hoa D. Nguyen, Bruce Benjamin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Obstructive sleep apnea (OSA) is a serious sleep disorder affecting millions of people worldwide. There is a great need to develop an efficient, low-cost OSA detection method. Traditional OSA detection methods are purely data-driven and hence their detection performance greatly depends on the quality and quantity of the sensor data. Several mathematical models of the human cardiorespiratory system exist which can generate different physiological signals that are hard to measure using current sensor technology. In this paper, we propose a new framework for OSA detection in which we fuse the sensor data with the physiological signal data from the cardiorespiratory system models. Multivariate Gaussian processes (GPs) are used to capture and model the physiological signal variations among different individuals. We define the multivariate GP covariance function using the sum of separable kernel functions form and estimate the corresponding hyperparameters by maximizing the GP marginal likelihood function. We detect OSA using the heart rate signal on a window-by-window basis using a likelihood ratio test. We conduct several experiments on both simulated and real data to show the effectiveness of the proposed OSA detection framework. We also compare with other purely data-driven OSA detection methods to demonstrate the advantage of the proposed OSA detection fusion framework.

Original languageEnglish
Title of host publicationFUSION 2016 - 19th International Conference on Information Fusion, Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages828-835
Number of pages8
ISBN (Electronic)9780996452748
StatePublished - 1 Aug 2016
Event19th International Conference on Information Fusion, FUSION 2016 - Heidelberg, Germany
Duration: 5 Jul 20168 Jul 2016

Publication series

NameFUSION 2016 - 19th International Conference on Information Fusion, Proceedings

Other

Other19th International Conference on Information Fusion, FUSION 2016
CountryGermany
CityHeidelberg
Period5/07/168/07/16

Fingerprint

Sleep
Data-driven
Model-based
Gaussian Process
Sensor
Sensors
Marginal Function
Marginal Likelihood
Hyperparameters
Covariance Function
Heart Rate
Electric fuses
Likelihood Ratio Test
Kernel Function
Likelihood Function
Disorder
Fusion
Fusion reactions
Mathematical Model
Mathematical models

Cite this

Gutta, S., Cheng, Q., Nguyen, H. D., & Benjamin, B. (2016). Model-based data-driven approach for sleep apnea detection. In FUSION 2016 - 19th International Conference on Information Fusion, Proceedings (pp. 828-835). [7527972] (FUSION 2016 - 19th International Conference on Information Fusion, Proceedings). Institute of Electrical and Electronics Engineers Inc..
Gutta, Sandeep ; Cheng, Qi ; Nguyen, Hoa D. ; Benjamin, Bruce. / Model-based data-driven approach for sleep apnea detection. FUSION 2016 - 19th International Conference on Information Fusion, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 828-835 (FUSION 2016 - 19th International Conference on Information Fusion, Proceedings).
@inproceedings{0e5039a3ee204432aac60af9cafab149,
title = "Model-based data-driven approach for sleep apnea detection",
abstract = "Obstructive sleep apnea (OSA) is a serious sleep disorder affecting millions of people worldwide. There is a great need to develop an efficient, low-cost OSA detection method. Traditional OSA detection methods are purely data-driven and hence their detection performance greatly depends on the quality and quantity of the sensor data. Several mathematical models of the human cardiorespiratory system exist which can generate different physiological signals that are hard to measure using current sensor technology. In this paper, we propose a new framework for OSA detection in which we fuse the sensor data with the physiological signal data from the cardiorespiratory system models. Multivariate Gaussian processes (GPs) are used to capture and model the physiological signal variations among different individuals. We define the multivariate GP covariance function using the sum of separable kernel functions form and estimate the corresponding hyperparameters by maximizing the GP marginal likelihood function. We detect OSA using the heart rate signal on a window-by-window basis using a likelihood ratio test. We conduct several experiments on both simulated and real data to show the effectiveness of the proposed OSA detection framework. We also compare with other purely data-driven OSA detection methods to demonstrate the advantage of the proposed OSA detection fusion framework.",
author = "Sandeep Gutta and Qi Cheng and Nguyen, {Hoa D.} and Bruce Benjamin",
year = "2016",
month = "8",
day = "1",
language = "English",
series = "FUSION 2016 - 19th International Conference on Information Fusion, Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "828--835",
booktitle = "FUSION 2016 - 19th International Conference on Information Fusion, Proceedings",

}

Gutta, S, Cheng, Q, Nguyen, HD & Benjamin, B 2016, Model-based data-driven approach for sleep apnea detection. in FUSION 2016 - 19th International Conference on Information Fusion, Proceedings., 7527972, FUSION 2016 - 19th International Conference on Information Fusion, Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 828-835, 19th International Conference on Information Fusion, FUSION 2016, Heidelberg, Germany, 5/07/16.

Model-based data-driven approach for sleep apnea detection. / Gutta, Sandeep; Cheng, Qi; Nguyen, Hoa D.; Benjamin, Bruce.

FUSION 2016 - 19th International Conference on Information Fusion, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2016. p. 828-835 7527972 (FUSION 2016 - 19th International Conference on Information Fusion, Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Model-based data-driven approach for sleep apnea detection

AU - Gutta, Sandeep

AU - Cheng, Qi

AU - Nguyen, Hoa D.

AU - Benjamin, Bruce

PY - 2016/8/1

Y1 - 2016/8/1

N2 - Obstructive sleep apnea (OSA) is a serious sleep disorder affecting millions of people worldwide. There is a great need to develop an efficient, low-cost OSA detection method. Traditional OSA detection methods are purely data-driven and hence their detection performance greatly depends on the quality and quantity of the sensor data. Several mathematical models of the human cardiorespiratory system exist which can generate different physiological signals that are hard to measure using current sensor technology. In this paper, we propose a new framework for OSA detection in which we fuse the sensor data with the physiological signal data from the cardiorespiratory system models. Multivariate Gaussian processes (GPs) are used to capture and model the physiological signal variations among different individuals. We define the multivariate GP covariance function using the sum of separable kernel functions form and estimate the corresponding hyperparameters by maximizing the GP marginal likelihood function. We detect OSA using the heart rate signal on a window-by-window basis using a likelihood ratio test. We conduct several experiments on both simulated and real data to show the effectiveness of the proposed OSA detection framework. We also compare with other purely data-driven OSA detection methods to demonstrate the advantage of the proposed OSA detection fusion framework.

AB - Obstructive sleep apnea (OSA) is a serious sleep disorder affecting millions of people worldwide. There is a great need to develop an efficient, low-cost OSA detection method. Traditional OSA detection methods are purely data-driven and hence their detection performance greatly depends on the quality and quantity of the sensor data. Several mathematical models of the human cardiorespiratory system exist which can generate different physiological signals that are hard to measure using current sensor technology. In this paper, we propose a new framework for OSA detection in which we fuse the sensor data with the physiological signal data from the cardiorespiratory system models. Multivariate Gaussian processes (GPs) are used to capture and model the physiological signal variations among different individuals. We define the multivariate GP covariance function using the sum of separable kernel functions form and estimate the corresponding hyperparameters by maximizing the GP marginal likelihood function. We detect OSA using the heart rate signal on a window-by-window basis using a likelihood ratio test. We conduct several experiments on both simulated and real data to show the effectiveness of the proposed OSA detection framework. We also compare with other purely data-driven OSA detection methods to demonstrate the advantage of the proposed OSA detection fusion framework.

UR - http://www.scopus.com/inward/record.url?scp=84992089268&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84992089268

T3 - FUSION 2016 - 19th International Conference on Information Fusion, Proceedings

SP - 828

EP - 835

BT - FUSION 2016 - 19th International Conference on Information Fusion, Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Gutta S, Cheng Q, Nguyen HD, Benjamin B. Model-based data-driven approach for sleep apnea detection. In FUSION 2016 - 19th International Conference on Information Fusion, Proceedings. Institute of Electrical and Electronics Engineers Inc. 2016. p. 828-835. 7527972. (FUSION 2016 - 19th International Conference on Information Fusion, Proceedings).