Description

This code uses the ActivityIndex R package to calculate Activity Index (AI), an acceleration-based metric which was compared to indirect calorimetry in a large sample of older women wearing an ActiGraph on their right hip in the laboratory (R²=0.72). Area under the curve from the receiver operating characteristic analysis was 0.93-0.96 for distinguishing sedentary, light, or moderate-to-vigorous physical activity. However, the provided code simply calculates AI at a user-specified epoch as specific cut-points are unavailable.

Development/Validation

Sample: 194 (all female) healthy adults, 60-91 years of age

Setting: Laboratory

Activities: Activities of daily living, overground/self-paced walking, treadmill walking, and stationary behaviors

Criterion: Oxycon Mobile (VO₂)

Accelerometer(s): ActiGraph GT3X+ on hip

Validation approach: Comparison with criterion

Phase Designation

(What's this?)

This model is in Phase 1.

Instructions

The R code can be used to load the R package that reads in a csv file (in one of two formats), calculates sigma (an intermediary needed for further calculations), then calculates Activity Index in the user-specified epoch.

There are two example data sheets- one is raw acceleration in each axis with no headers. The other is raw acceleration in the format exported by ActiLife- with 10 rows of header information. Further instructions are in the R code. More information about R is found here.

This code relies on R software which can be downloaded for free at https://www.r-project.org/

Attached Files

Download all files as a .zip or download individual files below.

Code

Example data with headers

Example data no headers

Note: Compared to the original source, code was edited for consistency, but sample data and package were unchanged

Source Information

Reference:

Bai J, Di C, Xiao L, Evenson KR, LaCroix AZ, Crainiceanu CM, et al. (2016) An Activity Index for Raw Accelerometry Data and its Comparison with Other Activity Metrics. PLoS ONE 11(8): e0160644. http://doi.org/10.1371/journal.pone.0160644 Link to paper

Original source for code:

https://github.com/javybai/ActivityIndex

Corresponding author: Jiawei Bai, jbai@jhsph.edu

Contact

Kimberly Clevenger at accelerometerrepository@gmail.com