Addition of chunk groups

Description

partition data into chunks of a fixed number of rows in order to calculate aggregated features per chunk

Usage

add_chunk_group(data, rows_per_chunk)

Arguments

Aggregate data into timesteps

Description

Aggregate data into timesteps

Aggregate E4 data into timesteps

Aggregate Embrace Plus data into timesteps

Aggregate Nowatch data into timesteps

Usage

aggregate_data(x, interval = "1 min")

aggregate_e4_data(x, interval = "1 min")

aggregate_embrace_plus_data(x, interval = "1 min")

aggregate_nowatch_data(x, interval = "1 min")

Arguments

as_time

Description

Converts Unix time to as.POSIXct

Usage

as_time(x, tz = "UTC")

Arguments

Convert an E4 data stream to a timeseries

Description

Creates an xts object indexed by time

Usage

as_timeseries(data, index = 2, name_col = "V1")

Arguments

Batch analysis

Description

Read and process all ZIP files in a directory

Usage

batch_analysis(path_in = NULL, path_out = ".")

Arguments

Configuration of the SVM algorithm for binary classification

Description

Configuration of the SVM algorithm for binary classification

Usage

binary_classifier_config

Format

An object of class list of length 4.

Author(s)

Sara Taylor sataylor@mit.edu

References

Taylor, S., Jaques, N., Chen, W., Fedor, S., Sano, A., & Picard, R. (2015). Automatic identification of artifacts in electrodermal activity data. In *Proceedings of the 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBS)*.

RMSSD calculation

Description

Calculation of RMSSD over 1 minute time periods for plotting

Usage

calculate_RMSSD(IBIdata)

Arguments

Force character datetime variable ("yyyy-mm-dd hh:mm:ss") to system timezone

Description

Force character datetime variable ("yyyy-mm-dd hh:mm:ss") to system timezone

Usage

char_clock_systime(time)

Arguments

Choice between two classes

Description

Make choice between two classes based on kernel values

Usage

choose_between_classes(class_a, class_b, kernels)

Arguments

Amplitude features

Description

Compute amplitude features.

Usage

compute_amplitude_features(data)

Arguments

Derivative features

Description

Compute derivative features.

Usage

compute_derivative_features(derivative, feature_name)

Arguments

Compute Features for SVM

Description

This function computes features for support vector machines (SVM) analysis. It processes input data, applies wavelet coefficients computation, and aggregates various statistical measures over defined time chunks. The function is designed to work with data frames containing electrodermal activity (EDA) data, filtered EDA, and timestamps.

Usage

compute_features2(data)

Arguments

Details

The function internally computes wavelet coefficients and then divides the data into chunks of specified durations (1 second, 0.5 seconds, and a custom duration based on amplitude features). For each chunk, it calculates various statistical measures. The resulting data frame includes timestamps and the computed features, which can be used for further analysis or as input to machine learning models like SVM.

Value

A data frame with computed features. Each row corresponds to a time chunk, and columns include statistical measures (like maximum, mean, standard deviation, median, and sum of positive values) of wavelet coefficients computed over different intervals (1 second, 0.5 seconds, and based on amplitude features).

Examples

## Not run: 
# Assuming 'my_data' is a preprocessed data frame with the necessary columns
features <- compute_features2(my_data)

## End(Not run)

Wavelet coefficients

Description

Compute wavelet coefficients.

Usage

compute_wavelet_coefficients(data)

Arguments

Wavelet decomposition

Description

Compute wavelet decomposition.

Usage

compute_wavelet_decomposition(data)

Arguments

Output folder

Description

Create output folder for E4 analysis results

Usage

create_e4_output_folder(obj, out_path = ".")

Arguments

Create an Empty Frequency List

Description

This function generates a list named 'freq' with predefined fields for frequency measurements. Most fields are initialized with NA values, except for 'type' which is set to "fourier".

Usage

create_empty_freq_list()

Value

A named list with frequency measurement fields, mostly set to NA.

Examples

empty_freq_list <- create_empty_freq_list()
str(empty_freq_list)

Create an Empty Time List

Description

This function generates a list named 'time' with predefined fields, all set to NA. The fields included are related to time-based measurements.

Usage

create_empty_time_list()

Value

A named list with all fields set to NA.

Examples

empty_time_list <- create_empty_time_list()
str(empty_time_list)

Small example dataset for e4

Description

Small example dataset for e4

Usage

e4_data

Format

An object of class "e4data"

Examples

data(e4_data)

Filter datasets for a Datetime start + end

Description

A function to determine how many intervals should be created. The question is at what time do you want the filecut to start, what should be the period that you want separate files for, and what should the interval be?

Usage

e4_filecut_intervals(time_start, time_end, interval)

Arguments

Function to filter the data object based on the time period and intervals that are needed for the files to be cut. The function also creates identical Empatica E4 zipfiles in the same directory as where the original zipfile is located.

Description

Function to filter the data object based on the time period and intervals that are needed for the files to be cut. The function also creates identical Empatica E4 zipfiles in the same directory as where the original zipfile is located.

Usage

filter_createdir_zip(
  data,
  time_start,
  time_end,
  interval,
  out_path = NULL,
  fn_name = NULL
)

Arguments

Value

out_path fn_name

Filter data according to a datetime start and end

Description

Filter data according to a datetime start and end

Filter all four datasets for a Datetime start + end

Filter all datasets for a Datetime start + end

Usage

filter_datetime(data, start, end)

filter_e4data_datetime(data, start, end)

filter_embrace_plus_data_timestamp(data, start, end)

Arguments

Function to find peaks of an EDA datafile

Description

This function finds the peaks of an EDA signal and adds basic properties to the datafile.

Usage

find_peaks(
  data,
  offset = 1,
  start_WT = 4,
  end_WT = 4,
  thres = 0.005,
  sample_rate = getOption("SAMPLE_RATE", 8)
)

Arguments

Details

Also, peak_end is assumed to be no later than the start of the next peak. Is that OK?

Value

data frame with several columns peaks 1 if apex peak_start 1 if start of peak peak_end 1 if end of preak peak_start_times if apex then corresponding start timestamp peak_end_times if apex then corresponding end timestamp half_rise if sharp decaying apex then time to halfway point in rise amp if apex then value of EDA at apex - value of EDA at start max_deriv if apex then max derivative within 1 second of apex rise_time if apex then time from start to apex decay_time if sharp decaying apex then time from apex to end SCR_width if sharp decaying apex then time from half rise to end

Peak width

Description

Get the width of the peak (in seconds, from halfway the rise until the end)

Usage

get_SCR_width(data, i_apex_with_decay)

Arguments

Peak amplitude

Description

Get the amplitude of the peaks

Usage

get_amp(data)

Arguments

Get the eda apex of the signal

Description

finds the apex of electrodermal activity eda signal within an optional time window

Usage

get_apex(eda_deriv, offset = 1)

Arguments

Decay time

Description

Get the time (in seconds) it takes to decay for each peak

Usage

get_decay_time(data, i_apex_with_decay)

Arguments

First derivative

Description

Get the first derivative.

Usage

get_derivative(values)

Arguments

Electrodermal activity signal derivative

Description

Finds the first derivatives of the eda signal

Usage

get_eda_deriv(eda)

Arguments

Half peak amp

Description

Get the amplitude value halfway between peak start and apex

Usage

get_half_amp(data, i)

Arguments

Half rise time

Description

Get the time (in seconds) it takes to get to halfway the rise in a peak

Usage

get_half_rise(data, i_apex_with_decay)

Arguments

Decaying peaks

Description

Identify peaks with a decent decay (at least half the amplitude of rise)

Usage

get_i_apex_with_decay(data)

Arguments

SVM kernel

Description

Generate kernel needed for SVM

Usage

get_kernel(kernel_transformation, sigma, columns)

Arguments

Maximum derivative

Description

Get the largest slope before apex, interpolated to seconds

Usage

get_max_deriv(data, eda_deriv, sample_rate)

Arguments

Peak end

Description

Find the end of the peaks, with some restrictions on the search

Usage

get_peak_end(data, max_lookahead)

Arguments

Peak end times

Description

Get the end timstamp of the peaks

Usage

get_peak_end_times(data)

Arguments

Start of peaks

Description

Provide info for each measurement whether it is the start of a peak (0 or 1)

Usage

get_peak_start(data, sample_rate)

Arguments

Peak start times

Description

Get the start times of the peaks

Usage

get_peak_start_times(data)

Arguments

Rise time of peaks

Description

Calculates the rise time of all peaks

Usage

get_rise_time(eda_deriv, apices, sample_rate, start_WT)

Arguments

Second derivative

Description

Get the second derivative.

Usage

get_second_derivative(values)

Arguments

IBI analysis version 2

Description

The 'ibi_analysis' function is an update to the previous 'ibi_analysis_old' (version 0.8.1) and includes improvements to handle Empatica E4 data more effectively. The updated version tries to mitigate issues with empty IBI files and reduces the likelihood of errors during HRV analysis.

Usage

ibi_analysis(IBI)

Arguments

Details

This function is an updated approach for the analysis of (IBI) data. Here is a link to the old version ibi_analysis_old. It was updated to handle specific errors encountered in the BuildNIHR or InterpolateNIHR functions that resulted from files that were almost empty on IBI data.

This function analyzes IBI data, focusing on time and frequency domain measures. Note that the function imports functions from the RHRV package for HRV (Heart Rate Variability) analysis. Also note that measurements on the wrist are now more often referred to as PRV (Pulse Rate Variability), see for instance https://jphysiolanthropol.biomedcentral.com/articles/10.1186/s40101-020-00233-x.

The function performs several steps in processing IBI data: 1. Initializes HRV data structure using RHRV. 2. Handles potential data discrepancies due to the nature of wrist-worn devices. 3. Implements error handling for BuildNIHR and InterpolateNIHR functions. 4. Processes the data through time and frequency domain analyses. 5. Returns a comprehensive summary of HRV measures.

Value

A list containing the results of the time analysis, frequency analysis (if available), and a summary of HRV measures. The summary includes time domain measures (SDNN, pNN50, etc.), frequency domain measures (HF, LF, LFHF, etc.), and counts of original and accepted beats.

Examples

## Not run: 
zip_path <- system.file("extdata", "1635148245_A00204.zip", package = "wearables")
Assuming "IBI_data" is your interbeat interval data
result <- read_e4("path to your file") 
print(result$IBI)

## End(Not run)

IBI analysis (Old version)

Description

This function is the older version of IBI (interbeat interval) analysis. It has been replaced by ibi_analysis. This version remains available for compatibility, but we want to deprecate the function in future versions of the package.

This function analyzes IBI data, focusing on time and frequency domain measures. Note that the function imports functions from the RHRV package for HRV (Heart Rate Variability) analysis. Also note that measurements on the wrist are now more often referred to as PRV (Pulse Rate Variability), see for instance https://jphysiolanthropol.biomedcentral.com/articles/10.1186/s40101-020-00233-x.

Usage

ibi_analysis_old(IBI)

Arguments

Details

The function performs several steps in analyzing IBI data: 1. Create an HRV data structure using RHRV. 2. Filter and preprocess the data, including handling of missing values and erroneous readings. 3. Perform time domain and frequency domain analyses. 4. Return a comprehensive list of results, including statistical summaries of the HRV measures.

Value

A list containing time analysis, frequency analysis (commented out in this version), and a summary of the results including time domain measures (SDNN, pNN50, etc.), frequency domain measures (HF, LF, LFHF, etc.), and the number of original and accepted beats.

Note

This function is deprecated and will be removed in future versions. Users should migrate to ibi_analysis for ongoing and future analyses.

See Also

ibi_analysis for the current version of IBI analysis.

Examples

## Not run: 
zip_path <- system.file("extdata", "1635148245_A00204.zip", package = "wearables")
Assuming "IBI_data" is your interbeat interval data
result <- read_e4("path to your file") 
print(result$IBI)

## End(Not run)

Join EDA Binary Classifier Output to Dataset

Description

This function joins the output of an EDA binary classifier to a dataset based on rounded 5-second intervals. It is designed to merge two data frames: one containing your main data and another containing EDA binary classifier predictions. The function ensures that each record in the main data is matched with the appropriate classifier output by aligning timestamps to the nearest 5-second interval.

Usage

join_eda_bin(data, eda_bin)

Arguments

Details

The function first uses 'padr::thicken' to extend the main data frame by creating a new column 'DateTime_5_sec', which rounds the 'DateTime' values to 5-second intervals. Then, it performs a left join with the EDA binary classifier data, which has been similarly adjusted using 'lubridate::floor_date' to match these intervals. After the join, unnecessary columns ('DateTime_5_sec' and 'id') are dropped, and the classifier's 'label' column is renamed to 'quality_flag'.

Value

A data frame that combines the main dataset with the EDA binary classifier outputs. The classifier's label is renamed to 'quality_flag'. In cases where a precise match for the 5-second interval is not found in the classifier output, NA values may be introduced in the 'quality_flag' column.

Examples

## Not run: 
  main_data <- data.frame(DateTime = as.POSIXct(...), ...)
  classifier_data <- data.frame(id = as.POSIXct(...), label = ..., ...)
  joined_data <- join_eda_bin(main_data, classifier_data)

## End(Not run)

Max value per segment of length n

Description

Give the maximum value of a vector of values per segment of length n.

Usage

max_per_n(values, n, output_length)

Arguments

Configuration of the SVM algorithm for ternary classification

Description

Configuration of the SVM algorithm for ternary classification

Usage

multiclass_classifier_config

Format

An object of class list of length 4.

Author(s)

Sara Taylor sataylor@mit.edu

References

Taylor, S., Jaques, N., Chen, W., Fedor, S., Sano, A., & Picard, R. (2015). Automatic identification of artifacts in electrodermal activity data. In *Proceedings of the 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBS)*.

pad_e4

Description

function to combine several e4 files, and sets the length of the x-axis

Usage

pad_e4(x)

Arguments

Artifact plots

Description

Plot artifacts after eda_data is classified

Usage

plot_artifacts(labels, eda_data)

Arguments

Binary classifiers

Description

Generate classifiers (artifact, no artifact)

Usage

predict_binary_classifier(data)

Arguments

Ternary classifiers

Description

Generate classifiers (artifact, unclear, no artifact)

Usage

predict_multiclass_classifier(data)

Arguments

prepend_time_column

Description

Column binds a time_column to the dataframe

Usage

prepend_time_column(data, timestart, hertz, tz = Sys.timezone())

Arguments

Show class of object

Description

Returns 'object of class'

Usage

## S3 method for class 'e4data'
print(x, ...)

Arguments

Process EDA data

Description

Process EDA data

Usage

process_eda(eda_data)

Arguments

Row-bind datasets

Description

Row-bind datasets

Usage

rbind_data(data)

Arguments

Details

Serves as an internal function that is called by 'rbind_e4', 'rbind_embrace_plus' and 'rbind_nowatch'. It takes a list of data.frames and row-binds them together.

Row-bind E4 datasets

Description

This function takes a list of E4 datasets and row-binds them together.

Usage

rbind_e4(data)

Arguments

Row-bind Embrace Plus datasets

Description

This function takes a list of Embrace Plus datasets and row-binds them together.

Usage

rbind_embrace_plus(data)

Arguments

Row-bind NOWATCH datasets

Description

This function takes a list of NOWATCH datasets and row-binds them together.

Usage

rbind_nowatch(data)

Arguments

Read, process and feature extraction of E4 data

Description

Reads the raw ZIP file using 'read_e4', performs analyses with 'ibi_analysis' and 'eda_analysis'.

Usage

read_and_process_e4(zipfile, tz = Sys.timezone())

process_e4(data)

process_embrace_plus(data)

Arguments

Value

An object with processed data and analyses, object of class 'e4_analysis'.

Read, process and feature extraction of Embrace Plus data

Description

Reads the raw ZIP file using 'read_embrace_plus', performs analyses with 'eda_analysis'.

Usage

read_and_process_embrace_plus(
  zipfile = NULL,
  folder = NULL,
  type = "raw",
  tz = Sys.timezone()
)

Arguments

Value

An object with processed data and analyses, object of class 'embrace_plus_analysis'.

Read E4 data

Description

Reads in E4 data as a list (with EDA, HR, Temp, ACC, BVP, IBI as dataframes), and prepends timecolumns

Usage

read_e4(zipfile = NULL, tz = Sys.timezone())

Arguments

Details

This function reads in a zipfile as exported by Empatica Connect. Then it extracts the zipfiles in a temporary folder and unzips the csv files in the temporary folder.

The EDA, HR, BVP, and TEMP csv files have a similar structure in which the starting time of the recording is read from the first row of the file (in unix time). The frequency of the measurements is read from the second row of the recording (in Hz). Subsequently, the raw data is read from row three onward.

The ACC csv file contain the acceleration of the Empatica E4 on the three axes x,y and z. The first row contains the starting time of the recording in unix time. The second row contains the frequency of the measurements in Hz. Subsequently, the raw x, y, and z data is read from row three onward.

The IBI file has a different structure, the starting time in unix is in the first row, first column. The firs column contins the number of seconds past since the start of the recording. The number of seconds past since the start of the recording represent a heartbeat as derived from the algorithms from the photo plethysmogrophy sensor. The second column contains the duration of the interval from one heartbeat to the next heartbeat.

ACC.csv = 32 Hz BVP.csv = 64 Hz EDA.csv = 4 HZ HR.csv = 1 HZ TEMP.csv = 4 Hz

Please also see the info.txt file provided in the zip file for additional information.

The function returns an object of class "e4_data" with a prepended datetime columns that defaults to user timezone. The object contains a list with dataframes from the physiological signals.

Examples

library(wearables)
# read_e4()

Read Embrace Plus data

Description

Reads in Embrace Plus data as a list (with EDA, HR, Temp, ACC, BVP, IBI as dataframes), and prepends timecolumns

Usage

read_embrace_plus(
  zipfile = NULL,
  folder = NULL,
  type = "raw",
  tz = Sys.timezone()
)

Arguments

Details

This function reads in a zipfile with data from the Embrace Plus device, or a folder with unzipped files. The unzipped files are avro or csv files.

The unzipped files are avro or csv files, where avro files are read in with using 'sparklyr', which sets up a local Spark cluster.

The function returns an object of class "embrace_plus_data" with a prepended datetime columns. The object contains a list with dataframes from the physiological signals.

Examples

## Not run: 
 library(wearables)
 read_embrace_plus(zipfile = "yourpathtohezipfile.zip")
 read_embrace_plus(folder = "/path/to/folder/with/files", type = "aggregated")

## End(Not run)

Read Nowatch data

Description

Reads in Nowatch data as a list, and prepends timecolumns

Usage

read_nowatch(zipfile = NULL, folder = NULL, tz = Sys.timezone())

Arguments

Details

This function reads in a zipfile with files exported by the Nowatch instrument, or a folder with the unzipped files. The files are expected to be csv files.

The unzipped files are csv files.

The function returns an object of class "nowatchdata" with a prepended datetime columns. The object contains a list with dataframes from the physiological signals.

Examples

## Not run: 
 library(wearables)
 read_nowatch("yourpathtohezipfile.zip")
 read_nowatch(folder = "/path/to/folder/with/files")

## End(Not run)

Small peaks removal

Description

Remove peaks with a small rise from start to apex are removed

Usage

remove_small_peaks(data, thres = 0)

Arguments

Upsample EDA data to 8 Hz

Description

Upsample EDA data to 8 Hz

Usage

upsample_data_to_8Hz(eda_data)

Arguments

Write CSV files of the output

Description

Slow!

Usage

write_processed_e4(obj, out_path = ".")

Arguments