Examining resting state EEG reliability between laboratory and clinical settings

Poster Session D - Monday, March 31, 2025, 8:00 – 10:00 am EDT, Back Bay Ballroom/Republic Ballroom

Erin S.M. Matsuba¹ (erin.matsuba@childrens.harvard.edu), Alex Job Said¹, Margaret Norberg¹, Charles A. Nelson¹, Carol L. Wilkinson¹; ¹Boston Children's Hospital

Electroencephalography (EEG) is increasingly being used as a tool to aid in early identification efforts for neurodevelopmental disorders. However, less is known regarding how EEG-based biomarkers may generalize from highly standardized laboratory settings into real-world environments. Thus, this study evaluates the reliability of resting state EEG features between laboratory and clinical settings. Resting state EEG recordings were obtained from participants aged 4 months to 34 years (n=16) across two settings: a space adjacent to a primary care waiting room of a large hospital and a dimly lit, electrically isolated, and sound-attenuated laboratory room. Mean absolute error (MAE) and intraclass correlation coefficients (ICC) were used to quantify the magnitude of difference between laboratory and clinic settings for the power spectrum, periodic and aperiodic components. The MAE between laboratory and clinical settings for the power spectrum was 0.12 (SD=0.07), the periodic component was 0.05 (SD=0.03), and the aperiodic component was 0.14 (SD=0.09). Previous studies have described acceptable MAE values for the adequacy of spectral fit between 0.025-0.1 (Ostlund et al., 2022). ICC (2,1) for periodic components was poor for delta (0.47), moderate for gamma (0.72) and high beta (0.75), and good for theta (0.76), low alpha (0.80), high alpha (0.83), and low beta (0.83) frequency bands. ICC (2,1) was good for aperiodic slope (0.81) and excellent for aperiodic offset (0.91). These findings suggest that resting-state EEG features demonstrate respectable reliability across settings. This increases the generalizability and accessibility of EEG as a potential clinical tool to identify neurodevelopmental disorders.

Topic Area: METHODS: Electrophysiology