acc (version 1.2.4)

acc: Summarizes accelerometer data for multiple types of physical activities

Description

Summarizes accelerometer for multiple types of physical activities

Usage

acc(data, tri, axis, spuriousDef, nonwearDef, minWear, patype, pacut, boutsize, tolerance)

Arguments

data

Data which consists of two columns [TimeStamp,counts] (i.e. accelerometer counts file read in by function readCounts)

tri

Whether the data is from a tri-axial accelerometer. Default is tri=`FALSE'. If tri=`TRUE' then option `axis' should be specified.

axis

This option is only used for the tri-axial accelerometer. Options are `x',`y',`z',`sum', or `vm'. Options `x', `y', or `z' can be specified to summarize data using only data from a single axis. If the option 'vm' is used, the square root of the squared sum of counts from three axes (i.e. $\sqrt{{x}^{2}+{y}^{2}+{z}^{2}}$) are used for the summary. If the option 'sum' is used, sum of the counts from three axes are used.

spuriousDef

Definition of spurious observation. Defined as minutes of consecutive zeros. For example, if spuriousDef = 20, this means that an observation point will be determined as a spurious observation if there are consequtive counts of at least 20 zeros before and after the single non-zero observation. Default is spuriousDef = 20.

nonwearDef

Definition of non-wear time. Defined as minutes of consecutive zeros. For example, if nonwearDef=60, this means that a period will be defined as non-wear time if there are at least 60 consecutive zeros. Default is nonwearDef=60. To consider all observations as wear time specify nonwearDef=`Inf'.

minWear

Minimum wear time definition. Defined as minutes of wear time. or example, if minWear = 600, this means that a day will be considered valid only if the wear time is at least 600 minutes. Default is minWear = 600. To return summary for all dates in the data, set minWear = 0.

patype

Types of physical activity for summary. For example, to summarize sedentary and moderate-vigorous physical activities, user specifies patype=c(`Sedentary',`MVPA'). This labels the summary accordingly.

pacut

Cut points to be used for the physical activity type. For example, if the user specified patype=c(`Sedentary',`MVPA'), pacut can be specified as pacut=c(c(0,99),c(1952,Inf)). The options requires to have a lower and a upper limit for each activity type (i.e. c(0,99) for sedentary activity). The specified interval includes its lower and upper endpoints (it is a closed interval).

boutsize

Boutsize to summarize a physical activity. If multiple patype is specified, boutsize should be specified for each one (e.g., if patype=c(`Sedentary',`MVPA') then one can use boutsize=c(10,10)).

tolerance

Whether two observations outside the physical activity should be permitted in summarizing a physical activity. If multiple patype is specified, tolerance should be for each one (e.g. if patype=c(`Sedentary',`MVPA') then one can use tolerance=c(`FALSE',`TRUE')).

Value

summary

Returns summary for each specified physical activity types (number of bouts and minutes of the activity), for valid dates.

References

Choi, L., Liu, Z., Matthews, C.E. and Buchowski, M.S. (2011). Validation of Accelerometer Wear and Nonwear Time Classification Algorithm. Med Sci Sports Exerc, 43(2):357-64.

Hall, K. S., Howe, C. A., Rana, S. R., Martin, C. L., and Morey, M. C. (2013). METs and Accelerometry of Walking in Older Adults: Standard versus Measured Energy Cost. Medicine and Science in Sports and Medicine, 45(3). 574-82.

Freedson, P., Melanson, E., and Sirard, J. (1998). Calibration of the Computer Sciences and Applications, Inc. accelerometer. Medicine and Science in Sports and Exercercise, 30(5):777-81.

Swartz, A. M., Strath, S. J., Bassett, D. R. Jr., O'Brien, W. L., King, G. A., and Ainsworth, B. E. (2000). Estimation of energy expenditure using CSA accelerometers at hip and wrist sites. Medicine and Science in Sports and Exercercise, 32: S450-456.

Copeland, J. L., and Esliger, D. W. (2009). Accelerometer assessment of physical activity in active, healthy older adults. J Aging Phys Act, 17: 17-30.

Examples

##
## Example: Loading accelerometer counts data using readCounts function
##      
## Not run: 
# library(acc)
# infile <- "CountsDataName.dat"
# counts <- readCounts(infile)
# ## End(Not run)

##
## Example: Summarizing accelerometer data for a sedentary individual
##        

# For this example, data is generated using a Hidden Markov model
# First, a sequence of time is generated
randomTime <- seq(ISOdate(2015,4,1),ISOdate(2015,4,3),"min")
# Load the mhsmm package to generate data using a Hidden Makov model
library(mhsmm)
# It is assumed that the counts are generated from a Hidden Markov model 
# with three states being non-wear, sedentary, and moderate-vigorous activity
J <- 3; initial <- rep(1/J, J)
# Set up a transition matrix for the Hidden Markov model.
P <- matrix(c(0.95, 0.04, 0.01, 
             0.09, 0.9, 0.01, 
             0.1, 0.2, 0.7), byrow='TRUE',nrow = J)
# It is assumed that the counts are realized from a mixture of
# two normal distributions (for sedentary activity and mvpa) 
# and a constant at zero (for non-wear time).
b <- list(mu = c(0, 30, 2500), sigma = c(0, 30, 1000))
model <- hmmspec(init = initial, trans = P, parms.emission = b,dens.emission = dnorm.hsmm)
# Generate data!
train <- simulate.hmmspec(model, nsim = (60*24*2), seed = 1234, rand.emis = rnorm.hsmm)
# Now set up a dataset that mimicks the accelerometry data
counts <- data.frame(TimeStamp = randomTime[1:length(train$x)], counts = train$x)
library(acc)
# summarize the data using the acc function.
# Sedentary and moderate-vigorous activity is summarized, using Freedson's cut points by default.
summary1 <- acc(data=counts, tri='FALSE', axis=NULL,
                     spuriousDef=20, nonwearDef=60, minWear=600, 
                     patype=c('Sedentary','MVPA'),pacut=c(c(0,99),c(1952,Inf)), 
                     boutsize=c(10,10), tolerance=c('FALSE','TRUE'))
summary1  


##
## Example: Summarizing accelerometer data for an active individual.
##

randomTime <- seq(ISOdate(2015,4,1),ISOdate(2015,4,3),"min")
library(mhsmm)
J <- 3; initial <- rep(1/J, J)
P <- matrix(c(0.95, 0.04, 0.01, 
             0.09, 0.7, 0.21, 
             0.1, 0.1, 0.8), byrow='TRUE',nrow = J)
b <- list(mu = c(0, 30, 2500), sigma = c(0, 30, 1000))
model <- hmmspec(init = initial, trans = P, parms.emission = b,dens.emission = dnorm.hsmm)
train <- simulate.hmmspec(model, nsim = (60*24*2), seed = 1234, rand.emission = rnorm.hsmm)

counts <- data.frame(TimeStamp = randomTime[1:length(train$x)], counts = train$x)
library(acc)
summary2 <- acc(data=counts, tri='FALSE', axis=NULL,
                     spuriousDef=20, nonwearDef=60, minWear=600, 
                     patype=c('Sedentary','MVPA'),pacut=c(c(0,99),c(1952,Inf)), 
                     boutsize=c(10,10), tolerance=c('FALSE','TRUE'))
summary2