RELAX-Jr: An Automated Pre-Processing Pipeline for Developmental EEG Recordings

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Manual electroencephalogram (EEG) data cleaning is time-consuming and prone to experimenter bias.
• Automated pipelines offer efficiency and objectivity but require adaptation for specific populations.
• The Reduction of Electroencephalographic Artefacts (RELAX) pipeline was previously developed for adult EEG data.

Purpose of the Study:

• To introduce and evaluate the RELAX-Jr pipeline, an automated system for pre-processing EEG data from children.
• To assess the performance of RELAX-Jr in identifying and reducing artifacts in pediatric EEG recordings.
• To compare RELAX-Jr's efficacy against other automated EEG cleaning methods.

Main Methods:

• RELAX-Jr utilizes multi-channel Wiener filtering (MWF) and/or wavelet-enhanced independent component analysis (wICA).
• It incorporates the adjusted-ADJUST algorithm for automated artifact classification.
• The pipeline was tested on resting-state EEG data from 136 children aged 4-12 years.

Main Results:

• RELAX-Jr demonstrated strong artifact reduction capabilities across various metrics.
• Performance was assessed using signal-to-error ratio, artifact-to-residue ratio, and reduction of blink/muscle artifacts.
• The pipeline showed effectiveness in preserving signal integrity for neurodevelopmental studies.

Conclusions:

• RELAX-Jr is a highly effective, fully automated pipeline for cleaning pediatric EEG data.
• The pipeline is suitable for neurodevelopmental research requiring high-quality EEG.
• RELAX-Jr reduces the need for manual intervention and expertise in EEG artifact removal.