ICLabel: An automated electroencephalographic independent component classifier, dataset, and website

By Luca Pion-Tonachini, Kenneth Kreutz-Delgado, Scott Makeig. NeuroImage, 2019

1,925 citations recorded by the source corpus.

The electroencephalogram (EEG) provides a non-invasive, minimally restrictive, and relatively low-cost measure of mesoscale brain dynamics with high temporal resolution. Although signals recorded in parallel by multiple, near-adjacent EEG scalp electrode channels are highly-correlated and combine signals from many different sources, biological and non-biological, independent component analysis (ICA) has been shown to isolate the various source generator processes underlying those recordings. Independent components (IC) found by ICA decomposition can be manually inspected, selected, and interpreted, but doing so requires both time and practice as ICs have no order or intrinsic interpretations and therefore require further study of their properties. Alternatively, sufficiently-accurate automated IC classifiers can be used to classify ICs into broad source categories, speeding the analysis of EEG studies with many subjects and enabling the use of ICA decomposition in near-real-time applications. While many such classifiers have been proposed recently, this work presents the ICLabel project comprised of (1) the ICLabel dataset containing spatiotemporal measures for over 200,000 ICs from more than 6,000 EEG recordings and matching component labels for over 6,000 of those ICs, all using common average reference, (2) the ICLabel website for collecting crowdsourced IC labels and educating EEG researchers and practitioners about IC interpretation, and (3) the automated ICLabel classifier, freely available for MATLAB. The ICLabel classifier improves upon existing methods in two ways: by improving the accuracy of the computed label estimates and by enhancing its computational efficiency. The classifier outperforms or performs comparably to the previous best publicly available automated IC component classification method for all measured IC categories while computing those labels ten times faster than that classifier as shown by a systematic comparison against other publicly available EEG IC classifiers.

Subfields: Cognitive Neuroscience, Signal Processing

View Paper: https://www.semanticscholar.org/paper/378402081d617c6c0ffb5cd3337d04b91cd74de4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Signal Processing

Citations in the corpus

Evaluation of Artifact Subspace Reconstruction for Automatic Artifact Components Removal in Multi-Channel EEG Recordings (2020, methodology): “ICA assumes EEG data, x, can be modeled as a linear mixture A of statistically independent sources, s, and learns an unmixing matrix, W , such that the independent components (IC) recover the original sources, y. x = As y = Wx ≈ s In this study, we employ extended Infomax ICA [37], which is availabl”
EEG Blink Artifacts Can Identify Read Music in Listening and Imagery (2025, unknown): “Independent component analysis (ICA) and ICLabel [22] were run to identify eye-related independent components (ICs).”
Novel Verbal Instructions Recruit Abstract Neural Patterns of Time-Variable Information Dimensionality (2025, methodology): “An Independent Component Analysis (ICA) was used to remove ocular (blink and saccades) and muscular artifacts identified through visual inspection and ICLabel (Pion-Tonachini et al., 2019).”
Age-related neural dynamics of empathy in preschoolers: an EEG study with naturalistic viewing (2026, unknown): “The ICLabel plugin (Pion-Tonachini et al. 2019) was used to classify components, and those identified as “Muscle,” “Eye,” “Heart,” “Line Noise,” or “Channel Noise” with > 80% probability were removed (11 participants excluded).”
Non-Neural Sources Systematically Impact Aperiodic EEG Activity (2026, unknown): “Probability scores assigned by ICLabel during preprocessing were used for classification.”
Is Limited Participant Diversity Impeding EEG-based Machine Learning? (2025, unknown): “There is a rich literature on EEG preprocessing [39, 40, 41], although most methods were designed for classical EEG analysis and it is unclear how well they translate to ML-based EEG applications.”
RELAX‐Jr: An Automated Pre‐Processing Pipeline for Developmental EEG Recordings (2024, unknown): “We note that within the original version of RELAX (designed for adult recordings) the wICA and ICA Subtract methods used ICLabel (Pion-Tonachini, Kreutz-Delgado, and Makeig 2019) to identify artefact components, while RELAX-Jr replaces ICLabel with adjusted-ADJUST (which is designed for use in paedi”
The development of oscillatory and aperiodic resting state activity is linked to a sensitive period in humans (2023, methodology): “Components representing artifacts were identified by employing the ICLabel classifier (Pion-Tonachini et al., 2019).”
Periodic and aperiodic neural activity displays age-dependent changes across early-to-middle childhood (2021, methodology): “Data were then subject to multiple Wiener filtering, followed by wavelet-enhanced ICA, with components for cleaning identified using IClabel (Pion-Tonachini et al., 2019).”
Contextual expectations in the real-world modulate low-frequency neural oscillations (2025, unknown): “The ICs identified by AMICA were copied into the original unfiltered dataset, and classified as “brain,” “eye,” “heart,” “channel noise,” “line noise,” “muscle,” and “other” using IClabel Lite (Pion-Tonachini et al., 2019).”
Neural oscillations guiding action during effects imagery. (2024, methodology): “To this end, we combined the IClabel algorithm [46] and dipole localization using the dipfit plugin to fit a dipole using the general boundary element head model to every IC using default settings according to the procedure suggested by [47].”
Task‐specific topology of brain networks supporting working memory and inhibition (2024, methodology): “IClabel (Pion-Tonachini et al., 2019) was used to classify the components into brain and non-brain activity.”