Adjusted-ADJUST

Adjusted-ADJUST automates identification and removal of artifacts from pediatric electroencephalography (EEG) data to improve retention and reliability of neural signals.

Key Features:

• Automated Artifact Identification: Optimizes the original ADJUST algorithm for pediatric EEG data collected with geodesic nets to distinguish artifactual from neural activity.
• Independent Component Analysis (ICA): Employs ICA to separate EEG signal components for targeted removal of movement and blink artifacts while preserving neural data.
• Performance Metrics: Evaluated by agreement with expert coders, number of trials retained after artifact removal, and reliability of the preprocessed EEG signal, with reported superior performance to the original ADJUST and ICLabel for pediatric datasets.

Scientific Applications:

• Pediatric EEG studies: Reduces the need for manual artifact identification and increases usable trial retention in infant, child, and other pediatric EEG datasets.
• Data quality improvement: Improves reliability of preprocessed EEG for downstream analyses by accurately classifying and removing artifacts from pediatric recordings.

Methodology:

Applies Independent Component Analysis (ICA); modifies the original ADJUST algorithm's artifact-detection criteria to account for age-specific EEG characteristics in pediatric and geodesic-net data; and validates efficacy via comparative analysis against original ADJUST and ICLabel across adult, child, and infant datasets.