P128 Physiologic artifacts when combining EEG and tDCS. Issue 3 (March 2017)

• Record Type: Journal Article
• Authors: Gebodh, N.; Adair, D.; Chelette, K.; Esmaeilpour, Z.; Bikson, M.; Dmochowski, J.
• Is Part Of: Clinical neurophysiology. Volume 128:Issue 3(2017:Mar.)
• Page Start: e78
• Publication Date: 2017-03
• Subjects: Neurophysiology -- Periodicals; Electroencephalography -- Periodicals; Electromyography -- Periodicals; Neurology -- Periodicals; 612.8
• DOI: 10.1016/j.clinph.2016.10.249
• Languages: English
• ISSNs: 1388-2457

Abstract

Abstract : The field of non invasive brain stimulation (NIBS) has benefited from integration with imaging including magnetic resonance imaging (MRI) and electroencephalography (EEG). Several studies have reported on concurrent tDCS and EEG, and used signal processing of varying complexity (e.g. high-pass filtering to ICA) to remove "non-physiologic stimulation artifacts" – namely artifacts arising from non-ideal stimulation and recording amplifier performance. None has addressed "physiologic artifacts" which are defined here as non-stationary changes in artifacts resulting from interactions between the stimulation induced voltage and body. We identified and systematically characterized a series of tDCS induced physiologic and non-physiologic artifacts during concurrent EEG and High Definition (HD)-tDCS, and adapted subject-specific computational modeling to corroborate physiological EEG findings. Physiologic artifacts include (1) cardiac distortion; (2) ocular motor distortion; (3) movement (myogenic) distortion. In each case, the artifact was montage, intensity, and polarity specific; as such contamination from these physiologic artifacts cannot be accounted for by typical control experiments (e.g. EEG changes that are dose specific). High resolution finite element models explained artifact based on specific impedance changes. Importantly (a) physiologic artifacts are universal, they are nominally independent of device and so exist regardless of devices; (b) the broad-band nature of contamination may confound a broad range of experiments (e.g. oscillations, ERP); (c) removal of artifacts requires recognition of their peculiar dynamic and individualized nature.