Resting state EEG classifies developmental status in three-year-old children

Area of Science:

• Neuroscience
• Developmental Psychology
• Biomedical Engineering

Background:

• Early childhood cognitive development monitoring is crucial for timely intervention.
• Current assessment methods are time-consuming and resource-intensive.
• Need for efficient, accessible tools for developmental screening.

Purpose of the Study:

• To evaluate the efficacy of short resting-state EEG (rs-EEG) in classifying developmental outcomes in 33-40-month-old children.
• To determine if rs-EEG can predict performance across multiple developmental domains.
• To assess feasibility in low-resource settings using portable EEG.

Main Methods:

• Collected 3-minute rs-EEG data from 70 children (33-40 months) using a 14-channel portable device.
• Applied supervised learning models using spectral and novel time-domain EEG features.
• Correlated EEG features with scores from the Bayley's Scale of Infant and Toddler Development, 3rd Edition (BSID-III).

Main Results:

• rs-EEG moderately classified five developmental domains (cognition, receptive/expressive language, fine/gross motor) with AUCs 0.70-0.84.
• Time-domain features showed stronger correlations and contributions than spectral frequencies.
• Accurate predictions were achievable with as few as 4 EEG channels.

Conclusions:

• Short rs-EEG is a viable, quick, and reliable indicator of cognitive developmental status in early childhood.
• This method can aid in identifying children needing developmental support, especially in low-resource environments.
• Portable, low-channel EEG systems show promise for accessible developmental screening.