Zero crossings for ActiGraph data

Description

Zero crossings for ActiGraph data

Usage

AG_zero_cross(data)

Arguments

Calculate proportion of zeroes in a data stream

Description

Calculate proportion of zeroes in a data stream

Usage

AG_zero_prop(data)

Arguments

ActiGraph sample data for exploring Sojourns Including Posture (SIP) method.

Description

ActiGraph sample data for exploring Sojourns Including Posture (SIP) method.

Usage

SIP_ag

Format

A data frame with 12257 rows and 5 variables:

counts

Activity counts from the first axis

axis2

Activity counts from the second axis

axis3

Activity counts from the third axis

vm

Vector magnitude of activity counts

Time

POSIX-formatted variable giving the timestamp for each observation

activPAL sample data for exploring Sojourns Including Posture (SIP) method

Description

activPAL sample data for exploring Sojourns Including Posture (SIP) method

Usage

SIP_ap

Format

A data frame with 12257 rows and 5 variables:

Time

POSIX-formatted variable giving the timestamp for each observation

DataCount

Integer value giving the index of the sample from which the row of data is drawn

Interval

Duration (in seconds) of the interval from one data point to the next

ActivityCode

Integer giving the posture activity classification: 0 is sedentary, 1 is standing, and 2 is stepping

CumulativeStepCount

Integer giving the cumulative step count from the start of the file to the current data point.

ActivityScore

Numeric giving MET-hours

Check the formatting of a data frame for use in the SIP pipeline

Description

Check the formatting of a data frame for use in the SIP pipeline

Usage

SIP_frame_test(frame, expected_var_names, expected_classes, verbose)

Arguments

Sojourn: Apply Sojourn Methods for Processing ActiGraph Accelerometer Data

Description

The Sojourn package provides a convenient way to apply the family of Sojourn methods published in previous works including Lyden K, Keadle S, Staudenmayer J, & Freedson P (2014) <doi:10.1249/MSS.0b013e3182a42a2d>, Ellingson LD, Schwabacher IJ, Kim Y, Welk GJ, & Cook DB (2016) <doi:10.1249/MSS.0000000000000915>, and Hibbing PR, Ellingson LD, Dixon PM, & Welk GJ (2018) <doi:10.1249/MSS.0000000000001486>.

Details

It is meant for use with data from ActiGraph monitors and (in the case of Sojourns Including Posture, by Ellingson et al. (2016)) activPAL monitors. File reading is not included in the functionality of the Sojourn package. For help with that preliminary step, users are directed to the packages AGread (for ActiGraph files) and activpalProcessing (for activPAL files).

Compute lag-one autocorrelation

Description

Compute lag-one autocorrelation

Usage

acf.lag1(x)

Arguments

Value

a numeric scalar giving the lag-one autocorrelation

Alternate lag-1 autocorrelation function

Description

This function is used to compute lag-1 autocorrelation for the SIP method

Usage

acf.lag1.alt(x)

Arguments

Apply the youth Sojourn method

Description

Function for using the youth Sojourn method developed by Hibbing et al. (2018)

Usage

apply_youth_sojourn(
  AG,
  vm = c("Vector.Magnitude", "ENMO"),
  Site = c("Hip", "Wrist"),
  demo_interactive = FALSE,
  verbose = FALSE,
  ...
)

Arguments

Value

The original data frame, plus additional predictions made by the Sojourn method

Note

The functions AGread::read_AG_counts and AGread::read_AG_raw are recommended for assembling the monitor-specific portion of the AG data frame.

Examples

data(example_data, package = "Sojourn")

  results_youth_soj <- apply_youth_sojourn(
    AG = example_data,
    vm = "Vector.Magnitude",
    Site = "Hip"
  )
  utils::head(results_youth_soj)

Combine short Sojourns

Description

Iteratively combine sojourns that are too short, until all Sojourns meet the minimum length requirement

Usage

combine_soj3x(durations, short, sojourns, verbose)

combine.sojourns(durations, short)

Arguments

Summarize outcomes from data processed using the Sojourn method

Description

A function to summarize predictions made by the original Sojourn method of Lyden et al. (2014).

Usage

compute.bouts.info(est.mets, units = c("secs", "mins"))

Arguments

Value

a data frame summarizing the predictions made by the Sojourn method.

Examples

data(example_data, package = "Sojourn")
example_data <- soj_3x_original(
  example_data$axis1,
  example_data$axis2,
  example_data$axis3,
  example_data$Vector.Magnitude
)

compute.bouts.info(example_data$METs)

Combine ActiGraph and activPAL data

Description

Merge data streams for separate monitors in the SIP method

Usage

enhance_actigraph(ag, ap, verbose = FALSE)

Arguments

Value

data frame of combined ActiGraph and activPAL data

Examples

data(SIP_ag, package = "Sojourn")
data(SIP_ap, package = "Sojourn")
combined_data <- enhance_actigraph(SIP_ag, SIP_ap)
utils::head(combined_data)

Sample data for exploring original and youth Sojourn methods

Description

Sample data for exploring original and youth Sojourn methods

Usage

example_data

Format

A data frame with 79989 rows and 11 variables:

id

An example ID for the data set

Sex

An example sex for the data set

Age

An example age (in years) for the data set

BMI

An example body mass index (in kg/m^2) for the data set

Timestamp

POSIX-formatted variable giving the timestamp for each observation

axis1

Activity counts from the first axis

axis2

Activity counts from the second axis

axis3

Activity counts from the third axis

Vector.Magnitude

Vector magnitude of activity counts (sqrt(sum(axis1^2, axis2^2, axis3^2)))

steps

Predicted steps taken

incline

Inclinometer status (0 = off, 1 = lying, 2 = sitting, 3 = standing)

Find transitions

Description

Examine a data stream for qualifying transitions according to the SIP method

Usage

find.transitions(y)

Arguments

Label Sojourns in a data stream according to the youth-specific algorithm

Description

Identify Sojourns using the algorithm of Hibbing et al. (2018)

Usage

get_youth_sojourns(
  vm,
  short = 30,
  Output = c("Counts", "Raw"),
  Site = c("Hip", "Wrist"),
  epoch = 1,
  difference = 15,
  threshold = 100,
  verbose = FALSE
)

Arguments

Value

A data frame (with nrow equal to length(vm)) that gives sojourn labels and durations

Examples

data(example_data, package = "Sojourn")
get_youth_sojourns(example_data$Vector.Magnitude,
  Output = "Counts", Site = "Hip")

Interactively input demographic information

Description

Interactively input demographic information

Usage

input_demographic(...)

Arguments

Value

A data frame containing an ID, attachment site of the monitor (hip or wrist), and the participant's sex, age, and BMI

Examples

if (interactive()) {
  input_demographic()
}

Function for tidy assembly of text for printing in the console

Description

Function for tidy assembly of text for printing in the console

Usage

messager(n, theGrid = NULL, id = "`Unspecified ID`")

Arguments

Value

text to print to the console

Shape data for input into the neural networks

Description

Prepare data for use in the neural networks of the SIP method

Usage

prep.nnetinputs(ag, sojourns, lag.fun)

Arguments

Prepare ActiGraph data for entry into the Sojourn neural network

Description

Prepare ActiGraph data for entry into the Sojourn neural network

Usage

prep_nnet_soj3x_original(
  df,
  variable = c("counts", "counts.2", "counts.3", "vect.mag"),
  suffix = ""
)

Arguments

Read an activPAL events file

Description

Read an activPAL events file

Usage

read_AP(filename, tz = "UTC")

Arguments

Value

Data frame reflecting the data contained in filename.

Note

There must be a corresponding .def file located in the same directory as filename

Examples

ap_file <- system.file(
"extdata/sampledata_Events.csv",
package = "Sojourn"
)
if (isTRUE(requireNamespace("data.table"))) {
  ap_data <- read_AP(ap_file)
  utils::head(ap_data)
}

Perform mean imputation

Description

This function is used to perform mean imputation for the SIP method

Usage

replace.na(x)

Arguments

Invoke the original uni-axial Sojourn method

Description

Calls the uni-axial Sojourn method from Lyden et al. (2014).

Usage

soj_1x_original(
  counts,
  perc.cut = 0.05,
  perc.cut.2 = 0.12,
  perc.cut.3 = 0.55,
  too.short = 10,
  sit.cut = 90,
  long.soj = 120
)

Arguments

Value

a data frame of processed data

Examples

data(example_data, package = "Sojourn")
results_1x <- soj_1x_original(example_data$axis1)
utils::head(results_1x)

Invoke the original triaxial Sojourn method

Description

Calls the triaxial Sojourn method from Lyden et al. (2014).

Usage

soj_3x_original(
  counts,
  counts.2,
  counts.3,
  vect.mag,
  short = 30,
  verbose = FALSE
)

Arguments

Value

a data frame of processed data

Examples

data(example_data, package = "Sojourn")
results_3x <- soj_3x_original(
  example_data$axis1,
  example_data$axis2,
  example_data$axis3,
  example_data$Vector.Magnitude
)

utils::head(results_3x)

Triaxial Sojourn method for the SIP method

Description

Triaxial Sojourn method for the SIP method

Usage

sojourn_3x_SIP(ag, short = 30)

Arguments

Value

A data frame of processed data using the SIP method

Examples

data(SIP_ag, package = "Sojourn")
data(SIP_ap, package = "Sojourn")
data <- Sojourn::enhance_actigraph(SIP_ag, SIP_ap)
utils::head(sojourn_3x_SIP(data))

Check monitor data for required demographic variables

Description

Check monitor data for required demographic variables

Usage

youth_name_test(
  AG,
  demo_names = c("id", "Age", "Sex", "BMI"),
  demo_interactive = FALSE,
  fix_sex_var = TRUE,
  ...
)

Arguments

Value

A properly-formatted data frame of monitor data

Format data for input into the youth Sojourn method's neural networks

Description

Format data for input into the youth Sojourn method's neural networks

Usage

youth_network_shape(
  data,
  sojourns = TRUE,
  RAW = FALSE,
  epoch = 1,
  verbose = TRUE,
  expand = TRUE,
  time_var = "Timestamp",
  first_print = TRUE,
  ...
)

Arguments

Value

A data frame containing data that are ready for making predictions with the youth Sojourn method's neural networks

Decision tree for making final predictions in the youth Sojourn method

Description

Decision tree for making final predictions in the youth Sojourn method

Usage

youth_sojourn_tree(
  Vector.Magnitude,
  Block.Est,
  Sojourns.Est,
  Hip = TRUE,
  RAW = FALSE,
  epoch = 1
)

Arguments

Value

A factor vector of predicted physical activity intensities (Sedentary, Light, or moderate-to-vigorous physical activity (MVPA))