A two-stage model for wearable device data

Research output: Contribution to journalArticle

Abstract

Recent advances of wearable computing technology have allowed continuous health monitoring in large observational studies and clinical trials. Examples of data collected by wearable devices include minute-by-minute physical activity proxies measured by accelerometers or heart rate. The analysis of data generated by wearable devices has so far been quite limited to crude summaries, for example, the mean activity count over the day. To better utilize the full data and account for the dynamics of activity level in the time domain, we introduce a two-stage regression model for the minute-by-minute physical activity proxy data. The model allows for both time-varying parameters and time-invariant parameters, which helps capture both the transition dynamics between active/inactive periods (Stage 1) and the activity intensity dynamics during active periods (Stage 2). The approach extends methods developed for zero-inflated Poisson data to account for the high-dimensionality and time-dependence of the high density data generated by wearable devices. Methods are motivated by and applied to the Baltimore Longitudinal Study of Aging.

Original languageEnglish (US)
JournalBiometrics
DOIs
StateAccepted/In press - 2017

Fingerprint

Two-stage Model
physical activity
Equipment and Supplies
Proxy
observational studies
Exercise
longitudinal studies
Accelerometers
Baltimore
heart rate
clinical trials
data analysis
Aging of materials
Wearable Computing
Health
Observational Studies
Longitudinal Studies
Observational Study
Time-varying Parameters
Monitoring

Keywords

  • Accelerometer
  • Actigraphy
  • Actiheart
  • Physical activity
  • Semi-parametric
  • Two-stage model

ASJC Scopus subject areas

  • Statistics and Probability
  • Biochemistry, Genetics and Molecular Biology(all)
  • Immunology and Microbiology(all)
  • Agricultural and Biological Sciences(all)
  • Applied Mathematics

Cite this

@article{4dde3cfb14ae4666b78e83c239ff5b86,
title = "A two-stage model for wearable device data",
abstract = "Recent advances of wearable computing technology have allowed continuous health monitoring in large observational studies and clinical trials. Examples of data collected by wearable devices include minute-by-minute physical activity proxies measured by accelerometers or heart rate. The analysis of data generated by wearable devices has so far been quite limited to crude summaries, for example, the mean activity count over the day. To better utilize the full data and account for the dynamics of activity level in the time domain, we introduce a two-stage regression model for the minute-by-minute physical activity proxy data. The model allows for both time-varying parameters and time-invariant parameters, which helps capture both the transition dynamics between active/inactive periods (Stage 1) and the activity intensity dynamics during active periods (Stage 2). The approach extends methods developed for zero-inflated Poisson data to account for the high-dimensionality and time-dependence of the high density data generated by wearable devices. Methods are motivated by and applied to the Baltimore Longitudinal Study of Aging.",
keywords = "Accelerometer, Actigraphy, Actiheart, Physical activity, Semi-parametric, Two-stage model",
author = "Jiawei Bai and Yifei Sun and Schrack, {Jennifer A} and Crainiceanu, {Ciprian M} and Wang, {Mei Cheng}",
year = "2017",
doi = "10.1111/biom.12781",
language = "English (US)",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",

}

TY - JOUR

T1 - A two-stage model for wearable device data

AU - Bai, Jiawei

AU - Sun, Yifei

AU - Schrack, Jennifer A

AU - Crainiceanu, Ciprian M

AU - Wang, Mei Cheng

PY - 2017

Y1 - 2017

N2 - Recent advances of wearable computing technology have allowed continuous health monitoring in large observational studies and clinical trials. Examples of data collected by wearable devices include minute-by-minute physical activity proxies measured by accelerometers or heart rate. The analysis of data generated by wearable devices has so far been quite limited to crude summaries, for example, the mean activity count over the day. To better utilize the full data and account for the dynamics of activity level in the time domain, we introduce a two-stage regression model for the minute-by-minute physical activity proxy data. The model allows for both time-varying parameters and time-invariant parameters, which helps capture both the transition dynamics between active/inactive periods (Stage 1) and the activity intensity dynamics during active periods (Stage 2). The approach extends methods developed for zero-inflated Poisson data to account for the high-dimensionality and time-dependence of the high density data generated by wearable devices. Methods are motivated by and applied to the Baltimore Longitudinal Study of Aging.

AB - Recent advances of wearable computing technology have allowed continuous health monitoring in large observational studies and clinical trials. Examples of data collected by wearable devices include minute-by-minute physical activity proxies measured by accelerometers or heart rate. The analysis of data generated by wearable devices has so far been quite limited to crude summaries, for example, the mean activity count over the day. To better utilize the full data and account for the dynamics of activity level in the time domain, we introduce a two-stage regression model for the minute-by-minute physical activity proxy data. The model allows for both time-varying parameters and time-invariant parameters, which helps capture both the transition dynamics between active/inactive periods (Stage 1) and the activity intensity dynamics during active periods (Stage 2). The approach extends methods developed for zero-inflated Poisson data to account for the high-dimensionality and time-dependence of the high density data generated by wearable devices. Methods are motivated by and applied to the Baltimore Longitudinal Study of Aging.

KW - Accelerometer

KW - Actigraphy

KW - Actiheart

KW - Physical activity

KW - Semi-parametric

KW - Two-stage model

UR - http://www.scopus.com/inward/record.url?scp=85031115052&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85031115052&partnerID=8YFLogxK

U2 - 10.1111/biom.12781

DO - 10.1111/biom.12781

M3 - Article

C2 - 29023644

AN - SCOPUS:85031115052

JO - Biometrics

JF - Biometrics

SN - 0006-341X

ER -