[Eeglablist] Electrical noise removal in intraoperative EEG data

Tarik S Bel-Bahar tarikbelbahar at gmail.com
Tue Mar 14 23:27:12 PDT 2017

Greetings Heiko, some notes below, best wishes.

Although you don't have many channels you have some options.

Programmatic comparisons of artifact rejection and noise cancellation
techniques in anesthesia have not yet been published.

Within eeglab lab there's a range of methods you could try, some are
listed below, plus other thoughts and possible resources.

You may benefit from looking at IEEE journals and clinical/medical
journals that also used only a few channels, selecting from the
best-of-breed methods. You may also want to google scholar some
articles (often from IEEE journals) that look at special denoising
techniques. See some examples at the end.

Though the ASR technique may not be built for only a few channels, you
may want to give a try, or a variant of it. In eeglab it exists as a
plugin for cleaning data, please google past eeglablist posts about
that. ASR might be able to detect cleaner periods (or feed the periods
to it) and then it may be able to "clean" the dirty periods. It does
this well with multi-channel EEG data.

Consider also  using the continuous (spectral) artifact detection tool
in eeglab , which should help you detect the worst periods even in a
single channel. You could also epoch a single channel or two into
periods, and run some of the eeglab epoch rejection tools.

To be safe, if you have an expert visual analyst of EEG data, they may
be able to help with a visual review to markup and "remove" bad

You may want to check out Cleanline plugin from Tim Mullen, which
works well for many.

Because you have a large sample, you should be able to estimate, in
some sub-sample of the participants, what really good data looks like,
and what really bad data looks like, and go from there.

Because your data is from frontal channels, it is likely permuted with
eye artifact, muscle artifact, and of course, our friend, the "true"
neural signal.

Note that multiple groups (e.g., Akeju et al., 2016) have published
results from ~4 channel frontal montages mostly likely similar to
yours. You may benefit most from mimicking their techniques.

Below are a selection of article titles that can be google scholared,
which you may find useful.

Please note that most/all methods in eeglab and and major eeg software
are multi-channel whole-head, and not usually built for "few" channel
analyses. From the looks of the below you have multiple options to

Electroencephalogram signatures of ketamine anesthesia-induced unconsciousness

O Akeju, AH Song, AE Hamilos, KJ Pavone… - Clinical …, 2016 - Elsevier

Automatic detection and classification of artifacts in single-channel EEG

T Olund, J Duun-Henriksen, TW Kjaer… - … in Medicine and …, 2014 -

Comparing the Performance of Popular MEG/EEG Artifact Correction
Methods in an Evoked-Response Study

NT Haumann, L Parkkonen, M Kliuchko… - … and Neuroscience, 2016 - dl.acm.org

Detection of eye blink artifacts from single prefrontal channel

WD Chang, HS Cha, K Kim, CH Im - Computer methods and programs in …,
2016 - Elsevier

Methods for artifact detection and removal from scalp EEG: A review

MK Islam, A Rastegarnia, Z Yang - Neurophysiologie Clinique/Clinical
…, 2016 - Elsevier

A comparison of five different algorithms for EEG signal analysis in
artifacts rejection for monitoring depth of anesthesia

Q Liu, YF Chen, SZ Fan, MF Abbod, JS Shieh - … Signal Processing and
…, 2016 - Elsevier

Comparative Study of Wavelet-Based Unsupervised Ocular Artifact
Removal Techniques for Single-Channel EEG Data

S Khatun, R Mahajan… - IEEE Journal of …, 2016 - ieeexplore.ieee.org

Removing Muscle Artifacts From EEG Data: Multichannel or
Single-Channel Techniques?

X Chen, A Liu, J Chiang, ZJ Wang… - IEEE Sensors …, 2016 - ieeexplore.ieee.org

EEG artifact removal—state-of-the-art and guidelines

JA Urigüen, B Garcia-Zapirain - Journal of neural engineering, 2015 -

Suppression of Eye-Blink Associated Artifact Using Single Channel EEG
Data by Combining Cross-Correlation With Empirical Mode Decomposition

R Patel, MPR Janawadkar, S Sengottuvel… - IEEE Sensors …, 2016 -

In-flight automatic detection of vigilance states using a single EEG channel

F Sauvet, C Bougard, M Coroenne… - IEEE Transactions …, 2014 -

Visualization of Whole-Night Sleep EEG From 2-Channel Mobile Recording
Device Reveals Distinct Deep Sleep Stages with Differential
Electrodermal …

JA Onton, DY Kang, TP Coleman - Frontiers in Human …, 2016 - ncbi.nlm.nih.gov

Signal processing techniques applied to human sleep EEG signals—A review

…, M Moshrefi-Torbati, M Hill, CM Hill, PR White - … Signal Processing
and …, 2014 - Elsevier

Monitoring burst suppression in critically ill patients: Multi-centric
evaluation of a novel method

F Fürbass, J Herta, J Koren, MB Westover… - Clinical …, 2016 - Elsevier

More information about the eeglablist mailing list