Signal Quality Investigation of a New Wearable Frontal Lobe EEG Device.

Zhilin Gao,Wang Wan,Zeguang Qin,Xingran Cui,Zhongze Gu

doi:10.3390/s22051898

Abstract

The demand for non-laboratory and long-term EEG acquisition in scientific and clinical applications has put forward new requirements for wearable EEG devices. In this paper, a new wearable frontal EEG device called Mindeep was proposed. A signal quality study was then conducted, which included simulated signal tests and signal quality comparison experiments. Simulated signals with different frequencies and amplitudes were used to test the stability of Mindeep’s circuit, and the high correlation coefficients (>0.9) proved that Mindeep has a stable and reliable hardware circuit. The signal quality comparison experiment, between Mindeep and the gold standard device, Neuroscan, included three tasks: (1) resting; (2) auditory oddball; and (3) attention. In the resting state, the average normalized cross-correlation coefficients between EEG signals recorded by the two devices was around 0.72 ± 0.02, Berger effect was observed (p < 0.01), and the comparison results in the time and frequency domain illustrated the ability of Mindeep to record high-quality EEG signals. The significant differences between high tone and low tone in auditory event-related potential collected by Mindeep was observed in N2 and P2. The attention recognition accuracy of Mindeep achieved 71.12% and 74.76% based on EEG features and the XGBoost model in the two attention tasks, respectively, which were higher than that of Neuroscan (70.19% and 72.80%). The results validated the performance of Mindeep as a prefrontal EEG recording device, which has a wide range of potential applications in audiology, cognitive neuroscience, and daily requirements.

Highlights

Advances in wearable EEG monitoring technology introduce new opportunities for EEG collection outside the laboratory and long-term monitoring in research and clinical areas [1]
This section included the results of the simulated signal test task and signal quality comparison experiments
A novel wearable frontal lobe EEG monitoring device Mindeep was proposed, and a rigorous signal quality evaluation analysis was performed on Mindeep

Summary

Introduction

Advances in wearable EEG monitoring technology introduce new opportunities for EEG collection outside the laboratory and long-term monitoring in research and clinical areas [1]. The accuracy of the portable EEG device B-Alert x10 (Advanced Brain Monitoring, ABM) [2] and in-ear EEG in emotional arousal have been shown to be 75.00% and 72.89%, and 71.21% and 71.07%. New wearable EEG devices, including Epitel Epilog [9], e-Glass [10], behind-the-ear EEG [11], Emotiv [12], and HealthSOS [13], have been shown to be useful for daily long-term monitoring of neurological diseases, such as epilepsy and stroke. Wearable EEG devices have the advantages of low power consumption, are light weight and portable, they are easy to operate, are low in price, and are more suitable for daily use and long-term monitoring in non-laboratory environments than traditional

Objectives

Methods

Results

Discussion

Conclusion