Eye Motion Detection using Electrooculography for Bio-signal Interfaces

• Published in: INTERNATIONAL JOURNAL OF AFFECTIVE ENGINEERING p. IJAE-D-24-00029
• Main Authors: TAMAKI, Daisuke, TANAKA, Hisaya
• Format: Journal Article
• Language: English
• Published: Japan Society of Kansei Engineering 2025
• Subjects: Active threshold; Electrooculography; Eye movement detection; Information transfer rate; k-Nearest neighbor
• ISSN: 2187-5413; 2187-5413
• DOI: 10.5057/ijae.IJAE-D-24-00029

This study aimed to determine the EOG interface that can be used for ALS, which can be used without preparation. Thus, eye movement detection was proposed using the RMS of the active threshold (AT) and K-NN methods. The AT method is dynamically performed using the RMS. In this study the combined use of both methods was described. Experiments on waveform detection accuracy were conducted on 19 healthy subjects between the ages of 20 and 29 years. The hit rate of the proposed method was 94%, and the false alarm rate was 9%. Next, we calculated the information transfer rate (ITR), a popular evaluation index for BCI. Using the proposed method, the ITR was bits/min. A P300 Speller is one of the bio-signal interfaces, that its ITR is 11 bits/min. Therefore, the ITR of the proposed method was higher than that of other bio-signal interfaces. Furthermore, the proposed method has shown great application potential as an interface.