Objective differentiation of neonatal EEG background grades using detrended fluctuation analysis

Area of Science:

• Computational Neuroscience
• Neonatal Neurology
• Signal Processing

Background:

• Objective assessment of neonatal electroencephalograph (EEG) background activity in sick infants is challenging.
• Traditional methods struggle with quantifying complex temporal correlations in neonatal EEG.
• Existing clinical gradings of EEG abnormalities lack quantitative precision.

Purpose of the Study:

• To investigate the utility of multifractal detrended fluctuation analysis (MF-DFA) for quantifying neonatal EEG background abnormalities.
• To determine if MF-DFA metrics can differentiate between various grades of EEG abnormality.
• To explore MF-DFA's potential for monitoring brain state changes during recovery.

Main Methods:

• Collected long-term EEG records from 34 neonates post-perinatal asphyxia.
• Applied MF-DFA to filtered (3-8 Hz) 15-minute EEG epochs.
• Compared MF-DFA metrics against visually assessed EEG grades and automated interburst interval detection.

Main Results:

• MF-DFA metrics showed significant differences between mild, moderate, and severe background EEG grades.
• MF-DFA parameters correlated significantly with automated interburst interval measurements.
• Piloted monitoring demonstrated MF-DFA metric evolution during patient recovery.

Conclusions:

• Neonatal EEG can be quantitatively assessed using multifractal metrics.
• MF-DFA offers a promising tool for grading EEG background abnormalities in neonates.
• This method may enable objective monitoring of brain state changes in long-term neonatal care.