Generate Activity Counts from Raw Accelerometer Data

A tool to obtain activity counts, originally a translation of the 'python' package 'agcounts' < https://github.com/actigraph/agcounts>. This tool allows the processing of data from any accelerometer brand, with a more flexible approach to handle different sampling frequencies.

actilifecounts

The goal of actilifecounts is to generate ActiGraph activity counts as described by Ali Neishabouri et al.

The actilifecounts R package is originally a translation of the python-based agcounts module.

Additional features of the actilifecounts package include the possibility of calculating activity counts from any accelerometer brand, with a more flexible approach to handle different sampling frequencies.

Installation

You can install the development version of actilifecounts from GitHub with:

# install.packages("devtools")
devtools::install_github("jhmigueles/actilifecounts")

Generating activity counts

This is a basic example which shows you how to generate activity counts.

library(actilifecounts)
# read accelerometer data (for example, from a csv file)
filename = "C:/mystudy/mydata/myfile.csv"
raw = data.table::fread(filename)
counts = get_counts(raw, sf = 100, epoch = 60, lfe_select = FALSE, verbose = TRUE)

If 3 axes are provided, the code assumes that the columns represent the X, Y, and Z axes in that order.