|
Activity Number:
|
177
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 1, 2016 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
|
Abstract #321395
|
|
|
Title:
|
A Two-Stage Model for Wearable Device Data
|
|
Author(s):
|
Jiawei Bai* and Yifei Sun and Ciprian Crainiceanu and Mei-Cheng Wang
|
|
Companies:
|
Johns Hopkins Bloomberg School of Public Health and The Johns Hopkins University and The Johns Hopkins University and The Johns Hopkins University
|
|
Keywords:
|
accelerometry ;
functional data ;
marked process ;
physical activity ;
wearable
|
|
Abstract:
|
Recent advances of wearable computing technology have allowed continuous health monitoring health studies. Data collected in such studies include minute-by-minute physical activity level provided by accelerometers and heart rate provided by heart rate monitors. However, the models and analyses of such data are very limited and mostly based on crude summaries. In this paper we will introduce a two-stage regression model that utilizes full minute-by-minute physical activity level data. This model is designed to explain both the transition dynamics between active/inactive periods (stage 1) and activity intensity dynamics during active periods (stage 2). The physical activity data from the Baltimore Longitudinal Study of Aging is used to illustrate the proposed methodology.
|
Authors who are presenting talks have a * after their name.
