accSummary: Summarizes accelerometer data for a single type of physical...
In acc: Exploring Accelerometer Data

Description Usage Arguments Value Author(s) References Examples

Summarizes accelerometer data for a single type of physical activity. Functionality is same as function acc, except that this function provides a summary for a single type of physical activity, with more detailed information.

1 2	accSummary(data, tri, axis, spuriousDef, nonwearDef, minWear, patype, pacut, epoch, boutsize, tolerance, returnbout)

`data`	Data which consists of two columns [TimeStamp,counts] (i.e. raw accelerometer file read in by function readRaw)
`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 spefied 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. √{{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`	Type of physical activity for summary. For example, to summarize sedentary activity, use option patype=c(‘Sedentary’). To summarize moderate-vigorous physical activities, user specifies patype=c(‘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’), pacut can be specified as pacut=c(c(0,99)). 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 inerval).
`boutsize`	Boutsize to summarize a physical activity. If multiple patype is specified, boutsize should be for each one (e.g., if patype=c(‘Sedentary’) then one can use boutsize=c(10)).
`epoch`	Epoch size. Default is '1 min'. Other epoch size can be specified using this option (e.g., '1 sec')
`tolerance`	Whether two observations outside the physical activity cut point should be permitted in summarizing a physical activity (e.g. if patype=c(‘Sedentary’) then one can use tolerance=c(‘FALSE’)).
`returnbout`	Whether to return data with bout indicators. If returnbout='FALSE' then only a summary of daily physical activity is returned.

If returnbout='FALSE', then a summary for each specified physical activity types (number of bouts and minutes of the activity), for valid dates. Defaults to 'TRUE'.

If returnbout='TRUE', then a list of summary object is returned with:

`totalDates`	Number of unique days available in data.
`validDates`	A summary for each specified physical activity types (number of bouts and minutes of the activity), for valid dates (as defined by minWear).
`PA`	Data supplied by the user is returned with three additional columns [inPA, nonwear, inboutPA], indicating whether the observation was considered to be in the defined cutpoint of the physical activity (inPA), in the defined cutpoint of nonwear time (nonwear), or in bout (inboutPA), respectively.

Jaejoon Song <jjsong2@mdanderson.org>

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.

##
## Example 1: Loading the activity counts data using readCounts function
##      
## Not run: 
library(acc)
infile <- "DataName.dat"
counts <- readCounts(infile)


##
## Example 2: 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.
# Option returnbout='TRUE' returns a more detailed information on how the summary was calculated.
summary1 <- accSummary(data=counts, tri='FALSE', axis=NULL,
                     spuriousDef=20, nonwearDef=60, minWear=600, 
                     patype='MVPA',pacut=c(1952,Inf), 
                     boutsize=10, tolerance='TRUE',returnbout='TRUE')
summary1$validDates # This returns the same summary as when returnbout='FALSE'
# summary1$PA # This returns the activity classification and bout information

##
## Example 3: 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)
# Option returnbout='TRUE' returns a more detailed information on how the summary was calculated.
summary2 <- accSummary(data=counts, tri='FALSE', axis=NULL,
                     spuriousDef=20, nonwearDef=60, minWear=600, 
                     patype='MVPA',pacut=c(1952,Inf), 
                     boutsize=10, tolerance='TRUE',returnbout='TRUE')
summary2$validDates # This returns the same summary as when returnbout='FALSE'
# summary2$PA # This returns the activity classification and bout information


## End(Not run)