ID 368 – Regional changes in detrend fluctuation analysis exponent values in healthy subjects

Abstract

Objective Detrended Fluctuation Analysis (DFA) was applied to EEGs recorded in healthy subjects as part of a study aimed at establishing the EEG in the bedside assessment of patients with disorder of consciousness (DOC). Analysis of normal human EEG indicates complex brain dynamics conforming to power laws statistics, with the suggestion of disrupted brain dynamics in DOC patients. Methods EEG recordings in healthy subjects ( n = 20),aged 24–59 years (mean 39.5, ± 10.11 years) were filtered between 0.5–70 Hz and sampled at 256 Hz. Artefact rejection was achieved by using independent component analysis (ICA). DFA Exponents were calculated for alpha (8–13 Hz), theta (4–7 Hz), beta (14–30 Hz) and delta (0.5–4 Hz) frequencies. Results Summarising exponent values across subjects indicates alpha DFA exponents were predictive for beta and theta DFA exponents values ( R 2 = 0.97 and 0. 93 respectively) and influenced by region. Conclusion Alpha DFA exponents were predictive for beta and theta exponent values suggesting an interaction between regions underpinning fluctuations of alpha, beta and theta oscillations. Applying this analysis to EEGs recorded in DOC patients could facilitate behavioural assessment and predict improved outcomes in the absence of unequivocal clinical signs of higher awareness. Key Message DFA exponents derived from EEG signals could facilitate characterization of patients with disorder of consciousness and equivocal clinical signs.