Dataset Information

Comparison of Medical and Consumer Wireless EEG Systems for Use in Clinical Trials.

ABSTRACT: Objectives: To compare quantitative EEG signal and test-retest reliability of medical grade and consumer EEG systems. Methods: Resting state EEG was acquired by two medical grade (B-Alert, Enobio) and two consumer (Muse, Mindwave) EEG systems in five healthy subjects during two study visits. EEG patterns, power spectral densities (PSDs) and test/retest reliability in eyes closed and eyes open conditions were compared across the four systems, focusing on Fp1, the only common electrode. Fp1 PSDs were obtained using Welch's modified periodogram method and averaged for the five subjects for each visit. The test/retest results were calculated as a ratio of Visit 1/Visit 2 Fp1 channel PSD at each 1 s epoch. Results: B-Alert, Enobio, and Mindwave Fp1 power spectra were similar. Muse showed a broadband increase in power spectra and the highest relative variation across test-retest acquisitions. Consumer systems were more prone to artifact due to eye blinks and muscle movement in the frontal region. Conclusions: EEG data can be successfully collected from all four systems tested. Although there was slightly more time required for application, medical systems offer clear advantages in data quality, reliability, and depth of analysis over the consumer systems. Significance: This evaluation provides evidence for informed selection of EEG systemsappropriate for clinical trials.

SUBMITTER: Ratti E

PROVIDER: S-EPMC5540902 | biostudies-other | 2017

REPOSITORIES: biostudies-other

ACCESS DATA

Similar Datasets

Project description:Recent advances in dry electrodes technology have facilitated the recording of EEG in situations not previously possible, thanks to the relatively swift electrode preparation and avoidance of applying gel to subject's hair. However, to become a true alternative, these systems should be compared to state-of-the-art wet EEG systems commonly used in clinical or research applications. In our study, we conducted a systematic comparison of electrodes application speed, subject comfort, and most critically electrophysiological signal quality between the conventional and wired Biosemi EEG system using wet active electrodes and the compact and wireless F1 EEG system consisting of dry passive electrodes. All subjects (n = 27) participated in two recording sessions on separate days, one with the wet EEG system and one with the dry EEG system, in which the session order was counterbalanced across subjects. In each session, we recorded their EEG during separate rest periods with eyes open and closed followed by two versions of a serial visual presentation target detection task. Each task component allows for a neural measure reflecting different characteristics of the data, including spectral power in canonical low frequency bands, event-related potential components (specifically, the P3b), and single trial classification based on machine learning. The performance across the two systems was similar in most measures, including the P3b amplitude and topography, as well as low frequency (theta, alpha, and beta) spectral power at rest. Both EEG systems performed well above chance in the classification analysis, with a marginal advantage of the wet system over the dry. Critically, all aforementioned electrophysiological metrics showed significant positive correlations (r = 0.54-0.89) between the two EEG systems. This multitude of measures provides a comprehensive comparison that captures different aspects of EEG data, including temporal precision, frequency domain as well as multivariate patterns of activity. Taken together, our results indicate that the dry EEG system used in this experiment can effectively record electrophysiological measures commonly used across the research and clinical contexts with comparable quality to the conventional wet EEG system.

Dataset Information

Comparison of Medical and Consumer Wireless EEG Systems for Use in Clinical Trials.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets