sleepeegpy

sleepeegpy is a high-level package built on top of mne-python, yasa and specparam (fooof) for preprocessing, analysis, and visualization of sleep EEG data.

Installation

0. Make sure you have Python version installed. Requires Python >3.9, <3.12.
1. Create a Python virtual environment, for more info you can refer to python venv, virtualenv or conda.
2. Activate the environment

3. pip install sleepeegpy
   

4. Download this repository zip folder, you will need only the notebooks folder.

Quickstart

1. Open the complete pipeline notebook in the created environment.
2. Follow the notebook's instructions.

RAM requirements

For overnight, high-density (256 channels) EEG recordings downsampled to 250 Hz expect at least 64 GB RAM expenditure for cleaning, spectral analyses and event detection.

Retrieve example dataset

odie = pooch.create(
    path=pooch.os_cache("sleepeegpy_dataset"),
    base_url="doi:10.5281/zenodo.10362189",
)
odie.load_registry_from_doi()
bad_channels = odie.fetch("bad_channels.txt")
annotations = odie.fetch("annotations.txt")
path_to_eeg = odie.fetch("resampled_raw.fif")
for i in range(1,4):
    odie.fetch(f"resampled_raw-{i}.fif")