Epilepsy Tools

This repository contains the epilepsy_tools Python package, which contains pre-written function to handle data from the various connected wearables used in the epilepsy research lab at CRCHUM.

Requirements

The package is built with Python 3.10 and newer in mind, and will not work with older versions. The list of required packages can be found in requirements.txt, and ahould be installed automatically when installing with the methods shown below

Installation

You can install from the repository directly, with the following command, with Git installed on your computer:

pip install git+https://github.com/CRCHUM-Epilepsy-Group/epilepsy_tools.git

You can also clone the repository or download it locally, and run the following command inside the root directory of the project:

pip install .

Quick Example

This is a quick example that loads data from a Cometa file (.c3d)

from epilepsy_tools import cometa

file_path = "some/directory/data.c3d

data = cometa.load_data(file_path)  # returns a pandas.DataFrame
recording_info = cometa.RecordingInfo.from_data(data)
emg_data = cometa.extract_emg_data(data)  # returns a pandas.DataFrame

Devices

This lists the current devices that have functions implemented for them.

Cometa

You can import the subpackage for the Cometa data with:

from epilepsy_tools import cometa

It uses the c3d package internally.

Functions:

• cometa.load_data(file): Load the data from a .c3d file. Returns a pandas.DataFrame.
• cometa.downsample(data, ratio): Returns a pandas.DataFrame with a lower frequency sampling.
• cometa.extract_emg_data(data): Returns only the EMG data from the provided pandas.DataFrame.
• cometa.extract_acceleration_data(data): Returns only the acceleration data from the provided pandas.DataFrame.
• cometa.plot_emg(data): Plot the EMG data from a Cometa DataFrame.
• cometa.plot_acceleration(data, *, norm=True): Plot the acceleration from a Cometa DataFrame. If norm=True (the default), calculate the norm of the acceleration vectors for each sensors (the X, Y and Z components) and plot that.

Classes:

• cometa.RecordingInfo: Metadata for a recording. Construct with cometa.RecordingInfo.from_file or cometa.RecordingInfo.from_data.

Constants:

• cometa.SENSOR_LABELS: A list of labels for the sensors currently used.

Hexoskin

You can import the subpackage for the Hexoskin data with:

from epilepsy_tools import hexoskin

It uses the pyEDFlib package internally.

Functions:

• hexoskin.load_data(file, *, as_dataframe=True): Load the data from a .edf file. Returns a pandas.DataFrame if as_dataframe is set to True (the default) or a dict of pandas.Series if set to False.

Classes:

• hexoskin.RecordingInfo: Metadata for a recording. Construct with hexoskin.RecordingInfo.from_file.

Logging

Enable logging for the module with:

import logging

logging.getLogger("epilepsy_tools").setLevel(logging.DEBUG)