Systematic Investigation of Optimal Electrode Positions and Re-Referencing Strategies on Ear Biosignals

Abstract

This study aims to determine the optimal electrode positions and re-referencing strategies for ear biosignals by systematically comparing ear and conventional biosignals. We analysed four physiological signals: electroencephalography (EEG), electromyography (EMG), electrooculography (EOG), and electrocardiography (ECG), with five re-referencing strategies. The signals were recorded from conventional locations and around the ears. Quantitative measures, such as the signal-to-noise ratio, F1 score, and correlation, were employed to identify the optimal electrode positions and re-referencing strategies for each physiological signal type. The optimal ear electrode positions were selected based on their proximity to conventional configurations: the upper part/behind the ear for EEG, the lower part for blink detection and vertical EOG, and the front for clench EMG and horizontal EOG. While no single re-referencing strategy consistently performed best for all physiological signals, we found that (mean-)ipsilateral re-referencing strategies were more suitable for EEG, EMG, blink detection, and vertical EOG, whereas (mean-)contralateral re-referencing strategies were more effective for horizontal EOG and ECG compared to other strategies. The optimal ear electrode positions and re-referencing methods varied slightly across subjects and demonstrated comparable performance to the conventional configurations in terms of quantitative measures. Ear biosignals have the potential for use in developing human-computer interfaces with comparable performance to conventional electrode configuration while being more practical in terms of attachment and detachment. This study provides in-depth insight into the optimal electrode locations and re-referencing strategy for developing ear-biosignal-based human-computer-interface applications.

Keywords:

• Brain-computer interface (BCI)
• ear biosignals
• human-computer interface (HCI)
• re-referencing strategy
• electroencephalography (EEG)
• electrooculography (EOG)
• electromyography (EMG)
• electrocardiography (ECG)

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data is available in https://doi.org/10.6084/m9.figshare.23523588.v1, and the code will be publicly published upon the acceptance.

Additional information

Funding

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) [No. RS-2023-00302489] and by the Bio & Medical Technology Development Program of the National Research Foundation (NRF) & funded by the Korean government (MSIT) [NRF-2021M3E5D2A01019547].

Notes on contributors

Seonghun Park

Seonghun Park is a postdoctoral researcher at Korea University, South Korea. He received his Ph.D. degree in Electronic Engineering from Hanyang University, South Korea. His current research interests include machine learning-based biomedical signal processing, and brain-computer interfaces in real-world scenarios.

Seong-Uk Kim

Seong-Uk Kim is a researcher at LG, South Korea. He received his master’s degree in Electronics and Information Engineering from Korea University, South Korea. His current research interests include biosignals-based sleep studies.

Soo-In Choi

Soo-In Choi is a postdoctoral researcher at the Korea Research Institute of Standards and Science, South Korea. She received her PhD degree in Medical IT Convergence Engineering from Kumoh National Institute of Technology. Her current research interests include brain-computer interfaces based on magnetoencephalography.

Han-Jeong Hwang

Han-Jeong Hwang is an associate professor at Korea University, South Korea. He received his PhD degree in Biomedical Engineering from Yonsei University, South Korea. His current research interests include machine learning-based biomedical signal processing, brain-computer interfaces, and neuromodulation.