Automatic detection of wakefulness and rest intervals in actigraphic signals: A data-driven approach

D. Martin-Martinez, Juan Pablo Casaseca, Jesus Maria Andres de Llano, J. R. Garmendia-Leiza, S. Alberola-Lopez, C. Alberola-Lopez

Research output: Contribution to journalArticle

Abstract

Actigraphy is an useful tool for evaluating the activity pattern of a subject; activity registries are usually processed by first splitting the signal into its wakefulness and rest intervals and then analyzing each one in isolation. Consequently, a preprocessing stage for such a splitting is needed. Several methods have been reported to this end but they rely on parameters and thresholds which are manually set based on previous knowledge of the signals or learned from training. This compromises the general applicability of this methods. In this paper we propose a new method in which thresholds are automatically set based solely on the specific registry to be analyzed. The method consists of two stages: (1) estimation of an initial classification mask by means of the expectation maximization algorithm and (2) estimation of a final refined mask through an iterative method which re-estimates both the mask and the classifier parameters at each iteration step. Results on real data show that our methodology outperforms those so far proposed and can be more effectively used to obtain derived sleep quality parameters from actigraphy registries. (C) 2014 IPEM. Published by Elsevier Ltd. All rights reserved.
Original languageEnglish
Pages (from-to)1585-1592
JournalMedical Engineering & Physics
Volume36
Issue number12
DOIs
Publication statusPublished - Dec 2014

Keywords

  • Actigraphy
  • Wakefulness/Rest detection
  • Probability density function
  • Expectation-maximization algorithm

Cite this

Martin-Martinez, D., Casaseca, J. P., Andres de Llano, J. M., Garmendia-Leiza, J. R., Alberola-Lopez, S., & Alberola-Lopez, C. (2014). Automatic detection of wakefulness and rest intervals in actigraphic signals: A data-driven approach. Medical Engineering & Physics, 36(12), 1585-1592. https://doi.org/10.1016/j.medengphy.2014.08.013
Martin-Martinez, D. ; Casaseca, Juan Pablo ; Andres de Llano, Jesus Maria ; Garmendia-Leiza, J. R. ; Alberola-Lopez, S. ; Alberola-Lopez, C. / Automatic detection of wakefulness and rest intervals in actigraphic signals: A data-driven approach. In: Medical Engineering & Physics. 2014 ; Vol. 36, No. 12. pp. 1585-1592.
@article{b967584f977c4ce292d0993e712b4d32,
title = "Automatic detection of wakefulness and rest intervals in actigraphic signals: A data-driven approach",
abstract = "Actigraphy is an useful tool for evaluating the activity pattern of a subject; activity registries are usually processed by first splitting the signal into its wakefulness and rest intervals and then analyzing each one in isolation. Consequently, a preprocessing stage for such a splitting is needed. Several methods have been reported to this end but they rely on parameters and thresholds which are manually set based on previous knowledge of the signals or learned from training. This compromises the general applicability of this methods. In this paper we propose a new method in which thresholds are automatically set based solely on the specific registry to be analyzed. The method consists of two stages: (1) estimation of an initial classification mask by means of the expectation maximization algorithm and (2) estimation of a final refined mask through an iterative method which re-estimates both the mask and the classifier parameters at each iteration step. Results on real data show that our methodology outperforms those so far proposed and can be more effectively used to obtain derived sleep quality parameters from actigraphy registries. (C) 2014 IPEM. Published by Elsevier Ltd. All rights reserved.",
keywords = "Actigraphy, Wakefulness/Rest detection, Probability density function, Expectation-maximization algorithm",
author = "D. Martin-Martinez and Casaseca, {Juan Pablo} and {Andres de Llano}, {Jesus Maria} and Garmendia-Leiza, {J. R.} and S. Alberola-Lopez and C. Alberola-Lopez",
year = "2014",
month = "12",
doi = "10.1016/j.medengphy.2014.08.013",
language = "English",
volume = "36",
pages = "1585--1592",
journal = "Medical Engineering & Physics",
issn = "1350-4533",
publisher = "Elsevier B.V.",
number = "12",

}

Martin-Martinez, D, Casaseca, JP, Andres de Llano, JM, Garmendia-Leiza, JR, Alberola-Lopez, S & Alberola-Lopez, C 2014, 'Automatic detection of wakefulness and rest intervals in actigraphic signals: A data-driven approach' Medical Engineering & Physics, vol. 36, no. 12, pp. 1585-1592. https://doi.org/10.1016/j.medengphy.2014.08.013

Automatic detection of wakefulness and rest intervals in actigraphic signals: A data-driven approach. / Martin-Martinez, D.; Casaseca, Juan Pablo; Andres de Llano, Jesus Maria; Garmendia-Leiza, J. R.; Alberola-Lopez, S.; Alberola-Lopez, C.

In: Medical Engineering & Physics, Vol. 36, No. 12, 12.2014, p. 1585-1592.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Automatic detection of wakefulness and rest intervals in actigraphic signals: A data-driven approach

AU - Martin-Martinez, D.

AU - Casaseca, Juan Pablo

AU - Andres de Llano, Jesus Maria

AU - Garmendia-Leiza, J. R.

AU - Alberola-Lopez, S.

AU - Alberola-Lopez, C.

PY - 2014/12

Y1 - 2014/12

N2 - Actigraphy is an useful tool for evaluating the activity pattern of a subject; activity registries are usually processed by first splitting the signal into its wakefulness and rest intervals and then analyzing each one in isolation. Consequently, a preprocessing stage for such a splitting is needed. Several methods have been reported to this end but they rely on parameters and thresholds which are manually set based on previous knowledge of the signals or learned from training. This compromises the general applicability of this methods. In this paper we propose a new method in which thresholds are automatically set based solely on the specific registry to be analyzed. The method consists of two stages: (1) estimation of an initial classification mask by means of the expectation maximization algorithm and (2) estimation of a final refined mask through an iterative method which re-estimates both the mask and the classifier parameters at each iteration step. Results on real data show that our methodology outperforms those so far proposed and can be more effectively used to obtain derived sleep quality parameters from actigraphy registries. (C) 2014 IPEM. Published by Elsevier Ltd. All rights reserved.

AB - Actigraphy is an useful tool for evaluating the activity pattern of a subject; activity registries are usually processed by first splitting the signal into its wakefulness and rest intervals and then analyzing each one in isolation. Consequently, a preprocessing stage for such a splitting is needed. Several methods have been reported to this end but they rely on parameters and thresholds which are manually set based on previous knowledge of the signals or learned from training. This compromises the general applicability of this methods. In this paper we propose a new method in which thresholds are automatically set based solely on the specific registry to be analyzed. The method consists of two stages: (1) estimation of an initial classification mask by means of the expectation maximization algorithm and (2) estimation of a final refined mask through an iterative method which re-estimates both the mask and the classifier parameters at each iteration step. Results on real data show that our methodology outperforms those so far proposed and can be more effectively used to obtain derived sleep quality parameters from actigraphy registries. (C) 2014 IPEM. Published by Elsevier Ltd. All rights reserved.

KW - Actigraphy

KW - Wakefulness/Rest detection

KW - Probability density function

KW - Expectation-maximization algorithm

U2 - 10.1016/j.medengphy.2014.08.013

DO - 10.1016/j.medengphy.2014.08.013

M3 - Article

VL - 36

SP - 1585

EP - 1592

JO - Medical Engineering & Physics

JF - Medical Engineering & Physics

SN - 1350-4533

IS - 12

ER -

Martin-Martinez D, Casaseca JP, Andres de Llano JM, Garmendia-Leiza JR, Alberola-Lopez S, Alberola-Lopez C. Automatic detection of wakefulness and rest intervals in actigraphic signals: A data-driven approach. Medical Engineering & Physics. 2014 Dec;36(12):1585-1592. https://doi.org/10.1016/j.medengphy.2014.08.013