XDAWN Denoising

XDAWN filters are trained from epochs, signal is projected in the sources space and then projected back in the sensor space using only the first two XDAWN components. The process is similar to an ICA, but is supervised in order to maximize the signal to signal + noise ratio of the evoked response.

WARNING: As this denoising method exploits the known events to maximize SNR of the contrast between conditions it can lead to overfit. To avoid a statistical analysis problem you should split epochs used in fit with the ones used in apply method.


[1] Rivet, B., Souloumiac, A., Attina, V., & Gibert, G. (2009). xDAWN algorithm to enhance evoked potentials: application to brain-computer interface. Biomedical Engineering, IEEE Transactions on, 56(8), 2035-2043.

[2] Rivet, B., Cecotti, H., Souloumiac, A., Maby, E., & Mattout, J. (2011, August). Theoretical analysis of xDAWN algorithm: application to an efficient sensor selection in a P300 BCI. In Signal Processing Conference, 2011 19th European (pp. 1382-1386). IEEE.

# Authors: Alexandre Barachant <alexandre.barachant@gmail.com>
# License: BSD (3-clause)

from mne import (io, compute_raw_covariance, read_events, pick_types, Epochs)
from mne.datasets import sample
from mne.preprocessing import Xdawn
from mne.viz import plot_epochs_image


data_path = sample.data_path()

Set parameters and read data

raw_fname = data_path + '/MEG/sample/sample_audvis_filt-0-40_raw.fif'
event_fname = data_path + '/MEG/sample/sample_audvis_filt-0-40_raw-eve.fif'
tmin, tmax = -0.1, 0.3
event_id = dict(vis_r=4)

# Setup for reading the raw data
raw = io.read_raw_fif(raw_fname, preload=True)
raw.filter(1, 20, method='iir')  # replace baselining with high-pass
events = read_events(event_fname)

raw.info['bads'] = ['MEG 2443']  # set bad channels
picks = pick_types(raw.info, meg=True, eeg=False, stim=False, eog=False,
# Epoching
epochs = Epochs(raw, events, event_id, tmin, tmax, proj=False,
                picks=picks, baseline=None, preload=True,
                add_eeg_ref=False, verbose=False)

# Plot image epoch before xdawn
plot_epochs_image(epochs['vis_r'], picks=[230], vmin=-500, vmax=500)

# Estimates signal covariance
signal_cov = compute_raw_covariance(raw, picks=picks)

# Xdawn instance
xd = Xdawn(n_components=2, signal_cov=signal_cov)

# Fit xdawn

# Denoise epochs
epochs_denoised = xd.apply(epochs)

# Plot image epoch after Xdawn
plot_epochs_image(epochs_denoised['vis_r'], picks=[230], vmin=-500, vmax=500)
  • ../../_images/sphx_glr_plot_xdawn_denoising_001.png
  • ../../_images/sphx_glr_plot_xdawn_denoising_002.png


Opening raw data file /home/ubuntu/mne_data/MNE-sample-data/MEG/sample/sample_audvis_filt-0-40_raw.fif...
    Read a total of 4 projection items:
        PCA-v1 (1 x 102)  idle
        PCA-v2 (1 x 102)  idle
        PCA-v3 (1 x 102)  idle
        Average EEG reference (1 x 60)  idle
    Range : 6450 ... 48149 =     42.956 ...   320.665 secs
Current compensation grade : 0
add_eeg_ref defaults to True in 0.13, will default to False in 0.14, and will be removed in 0.15. We recommend to use add_eeg_ref=False and set_eeg_reference() instead.
Reading 0 ... 41699  =      0.000 ...   277.709 secs...
Band-pass filtering from 1 - 20 Hz
Using up to 1388 segments
Number of samples used : 41640
Transforming to Xdawn space
Zeroing out 303 Xdawn components
Inverse transforming to sensor space

Total running time of the script: ( 0 minutes 5.025 seconds)

Generated by Sphinx-Gallery