Abstracts

Rationale: EEG is a gold standard for epilepsy diagnosis. Nevertheless, bottlenecks for reaching its full potential in neurology practice include the shortage of trained EEG technologists and the technical skills of manual electrode setup (i.e., marking locations, preparing skin, applying conductive paste, etc) to ensure consistent signal quality. Following investment from an Epilepsy Foundation grant, Zeto (Santa Clara, CA) has developed a 19-channel, wireless, zero-preparation, adjustable headset (zEEG) equipped with replaceable magnetic tip, dry electrodes (Figures1). zEEG is uploaded real time via Wi-Fi to cloud servers and data can be viewed/analyzed real time from any location. To validate the effectiveness of zEEG as a routine EEG, we performed recordings of a reference EEG system and zEEG at an outpatient clinic. We compared the subjective quality of recordings from a sample of patients from a tertiary referral epilepsy clinic. Methods: All studies were acquired by registered EEG technologists. The zEEG headset was placed using the manufacturer’s recommended procedure. The headset was removed and then a standard, routine EEG was acquired with a reference system (Natus, Pleasanton, CA). Standard cup electrodes were placed according to the 10-20 system using standard techniques.EEGs were recorded for 20-30 minutes. EEGs were then exported to common review software (Persyst, Solana Beach, CA) and reviewed in a longitudinal bipolar montage using set viewing settings (TC 0.16s, LFF 70Hz, notch 60Hz). 4 board-eligible or -certified reviewers blinded to each other’s ratings and machine source rated EEGs according to subjective 3 point Likert scale (3=highest quality, 2=acceptable, 1=not acceptable, 0 = n/a) in the following categories: myogenic noise, body movements, baseline wander and slow frequency artifact, electrode defect, 60 Hz artifact, waking alpha rhythm, and other waking background activities. Scores were averaged across reviewers and studies within each category were compared by recording system using one-sided Wilcoxon tests to evaluate the hypothesis that zEEG was non-inferior to the reference EEG system. Results: Results are detailed in Table 1. Electrode defects, baseline wander and slow frequency artifacts, and body movements were significantly more prominent in the Zeto system. Myogenic noise artifact, 60 Hz electrical artifact, the waking posterior rhythm, and other waking background activities were not significantly different compared to that of the reference system. Conclusions: Electrode artifacts and slow frequency artifacts are more apparent in the Zeto system but the overall interpretability is not inferior to that of the reference system. Overall, dry electrode systems can provide clinically relevant routine EEG recordings. Funding: Zeto Inc. provided funds to complete this study.