Research

Tracking vigilant attention using eye metrics

We have developed a technology that can accurately estimate mild decreases in alertness—something that has been difficult to achieve with conventional drowsiness assessment methods—using only eyelid-related indices measurable through non-contact eye cameras or webcams. With this technology, we aim to establish a method for detecting mild drowsiness during driving. Furthermore, we seek to develop a non-intrusive alertness assessment tool for various populations, including e-sports athletes and office workers. By leveraging this technology, we will advance research aimed at understanding and preventing decreases in alertness, contributing to the realization of a safer and more productive society.

Related Papers

  • Abe, T. (2023) PERCLOS-based technologies for detecting drowsiness: current evidence and future directions, SLEEP Advances, Volume 4, Issue 1, zpad006, https://doi.org/10.1093/sleepadvances/zpad006
  • Tanguchi, K*, Noguchi, T.*, Iizuka, S., Ando, H. Abe, T. and Fukui, K. (2023) A New Large-Scale Video Dataset of the Eyelid Opening Degree for Deep Regression-based PERCLOS Estimation. 26th International Conference on Medical Image Computing and Computer Assisted Intervention
  • Abe,T., Mishima, K., Kitamura, S., Hida, A., Inoue, Y., Mizuno, K., Kaida, K., Nakazaki, K.,Motomura, Y., Maruo, K., Ohta, T., Furukawa, S., Dinges, D.F., Ogata, K. (2020). Tracking Intermediate Performance of Vigilant Attention Using Multiple Eye Metrics. Sleep, 43, zsz219.
  • Abe, T., Nonomura, T., Komada, Y., Asaoka, S., Sasai, T., Ueno, A., Inoue, Y. (2011). Detecting deteriorated vigilance using percentage of eyelid closure time during behavioral maintenance of wakefulness tests. International Journal of Psychophysiology. 82, 269-274.

Decoding dream emotions using electroencephalography

We are developing a technology that can accurately estimate the emotional content of dreams during REM sleep using EEG, aiming to establish a "no-report paradigm for dream emotions" that enables objective evaluation without relying on subjective reports. Using this technology, we aim to elucidate the functions of emotions during REM sleep.

Related Papers

  • Luis Alfredo Moctezuma*, Marta Molinas, Takashi Abe. Unlocking Dreams and Dreamless Sleep: Machine Learning Classification with Optimal EEG Channels. BioMed Research International
  • Luis Alfredo Moctezuma, Kazuki Sato, Marta Molinas, and Takashi Abe (2023). Decoding emotion dimensions arousal and valence elicited on EEG responses to videos and images: a comparative evaluation. Brain Informatics 2023
  • Luis Alfredo Moctezuma, Felix Ipanaque, Marta Molinas, Takashi Abe. (2023). Dream emotions identified without awakenings by machine and deep learning from Electroencephalographic signals in REM sleep. Proceedings of 2023 IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering.
  • Moctezuma, L. A.*, Abe, T., Molinas, M. (2022). Two-dimensional CNN-based distinction of human emotions from EEG channels selected by multi-objective evolutionary algorithm. Scientific Reports, 12(1), 3523.

Improving sleep through non-invasive intervention

We aim to understand the phenomenon in which monotonous and rhythmic sensory stimulation facilitates sleep onset, and to develop a non-invasive, non-pharmacological method to promote sleep.

Related Papers

  • Zhiwei Fan, Yunyao Zhu, Chihiro Suzuki, Yoko Suzuki, Yumi Watanabe, Junki Endo, Takahiro Watanabe, Takashi Abe (under review) Binaural Beats at 0.25 Hz Shorten Latency to Slow-Wave Sleep during Daytime Nap.