This is a basic implementation of functionality provided in the open-source R package {mnirs}.

Available from: https://github.com/jemarnold/mnirs

For more information see the {mnirs} package documentation: https://jemarnold.github.io/mnirs/index.html

Author: Jem Arnold

mNIRS files can be imported and processed using standardised methods, and displayed in a plot and data table. The plot is interactive and can be zoomed in and out (changing settings will reset plot zoom). Processed data can be downloaded for further analysis.

Upload File:

Upload an .xls(x) or .csv file containing mNIRS data exported from many common wearable devices. Files exported from common NIRS devices should be automatically recognised, with the first detected nirs_channel returned.

mNIRS Channel Names:

Specify the column name(s) containing mNIRS data. Multiple channels can be specified using comma-separated new_name = file_column_name pairs.

Example: smo2_left = SmO2, smo2_right = SmO2(2)

Time/Sample Channel Name:

Specify the column containing time or sample values.

Example: time = Timestamp (seconds)

Lap/Event Channel Name (optional):

Optionally specify column with lap/event markers.

Zero Start Time:

Reset time_channel to start at zero.

Sample Rate:

Specify the exported data sample rate in Hz. This will be automatically estimated from the time channel and can be manually overridden.

Resample Rate:

Data can be resampled to a higher or lower sample rate. Also used to correct time_channel values for data with irregular or duplicated samples.

Trim Head/Tail Timespan:

Remove samples from the beginning or end of the recording, specified in units of time_channel (i.e., seconds).

Replace Invalid Values:

Replace specific fixed values (e.g., c(0, 100)) from nirs_channels.

Replace Outliers:

Remove local outliers using a moving window Hampel filter approach.

Replace Missing Values:

Linearly interpolate across missing values in nirs_channel.

Digital Filter Method:

Apply smoothing filters to improve signal-to-noise ratio. Methods include a cubic "smoothing-spline", a low-pass "Butterworth" filter, or a simple "moving average". Additional parameters for each filter method can be specified.

Shift Data:

Move nirs_channels values up or down to a new specified reference value, based on the "first", "minimum", or "maximum" data points. Multiple channels can be shifted together ("Ensemble") or independently ("Distinct").

Rescale Data:

Normalise nirs_channels to a new specified dynamic range. Multiple channels can be shifted together ("Ensemble") or independently ("Distinct").

Place Event Markers:

Manually add event markers at specified time points. Will add an event_channel to the data table if not already specified, otherwise will add character event labels or numeric time values to an existing event_channel in the appropriate format.

Keep All Columns:

Either keep all columns in the file data table (the default), or keep only the channels specified.

Download Data:

Export processed data as an Excel file for further analysis.