On Utilizing Energy Expenditure in Estimating Health Levels from Raw Mobility Data

Advisor Information

Hesham Ali

Location

UNO Criss Library, Room 232

Presentation Type

Oral Presentation

Start Date

4-3-2016 2:00 PM

End Date

4-3-2016 2:15 PM

Abstract

The general correlation between mobility patterns and health conditions has been established by many research studies. However, not much has been reported in terms of how to utilize such relationship for objectively assess health levels or use mobility data to predict health hazards. Although many instruments such as accelerometers are currently available for collecting mobility data, there is still a big gap between the collection of raw mobility data and the ability to obtain useful knowledge that leads to informed decisions through in-depth analysis of such data. In addition, not all reported results are homogeneous in terms of using similar mobility parameters and/or exact experimental conditions, which adds another level of difficulty in order to aggregate raw data and reach statistically significant levels. In this study, we propose the use of Energy Expenditure (EE) as a unifying parameter to convert various raw accelerometer data to common data points ready for aggregation and analysis. Although some devices are able to estimate the amount of Energy spent during various activities, the estimated values are not produced with the level of accuracy needed for the in-depth analysis required to predict health levels. In addition, it has been reported that estimating energy could be highly dependent on the location of the monitoring device. Few studies even attempted to estimate the energy associated with accelerometer data collected when the device is located on wrists or hips. In this research, we conduct a comprehensive analysis to study the impact of accelerometer position on the estimation of expended energy. We also develop regression model predict EE associated with raw accelerometer data. In particular, we are interested in looking at the correlation between Energy Expenditure and mobility data acquired from various locations such ankle or hip.

This document is currently not available here.

COinS
 
Mar 4th, 2:00 PM Mar 4th, 2:15 PM

On Utilizing Energy Expenditure in Estimating Health Levels from Raw Mobility Data

UNO Criss Library, Room 232

The general correlation between mobility patterns and health conditions has been established by many research studies. However, not much has been reported in terms of how to utilize such relationship for objectively assess health levels or use mobility data to predict health hazards. Although many instruments such as accelerometers are currently available for collecting mobility data, there is still a big gap between the collection of raw mobility data and the ability to obtain useful knowledge that leads to informed decisions through in-depth analysis of such data. In addition, not all reported results are homogeneous in terms of using similar mobility parameters and/or exact experimental conditions, which adds another level of difficulty in order to aggregate raw data and reach statistically significant levels. In this study, we propose the use of Energy Expenditure (EE) as a unifying parameter to convert various raw accelerometer data to common data points ready for aggregation and analysis. Although some devices are able to estimate the amount of Energy spent during various activities, the estimated values are not produced with the level of accuracy needed for the in-depth analysis required to predict health levels. In addition, it has been reported that estimating energy could be highly dependent on the location of the monitoring device. Few studies even attempted to estimate the energy associated with accelerometer data collected when the device is located on wrists or hips. In this research, we conduct a comprehensive analysis to study the impact of accelerometer position on the estimation of expended energy. We also develop regression model predict EE associated with raw accelerometer data. In particular, we are interested in looking at the correlation between Energy Expenditure and mobility data acquired from various locations such ankle or hip.