2024.10.24
Dr. Luis Alfredo Moctezuma (Abe laboratory) received the “Best Poster Presentation Award” at the Asian Society of Sleep Medicine Interim Meeting 2024, held in Tokyo, Japan on October 18-19, 2024.
Dr. Moctezuma, a JSPS postdoctoral researcher at the International Institute for Integrative Sleep Medicine, was recognized for his outstanding presentation titled “Automated Dream Detection in NREM Sleep Stage 2 Using High- and Low-Density EEG.”
He demonstrated how electroencephalographic (EEG) signals and machine learning could be utilized to automatically identify and differentiate between dream experience (DE) and no experience (NE) during the sleep stage N2 of non-rapid eye movement sleep. His results showed that it is possible to obtain up to 94% in accuracy, F1-score, Precision, Recall, AUROC of 97.5%, and a kappa of 0.88. Furthermore, he used channel selection algorithms to identify the most relevant EEG channels, and although performance metrics decreased as the number of EEG channels was reduced, performance remained similar using up to 30-40 EEG channels. He demonstrates high classification performance for automatic dream detection and informs the minimum required number of EEG channels to create machine learning models, thus obtaining low-cost portable devices that can be used in real-life scenarios.
Machine learning models created to classify NE and DE using high-density EEG could identify if a subject was having a dream with an accuracy of 70% even when subjects reported a dream experience without recall.
Dr. Moctezuma expressed his gratitude for the recognition, stating, “I am truly honored to receive this award and feel incredibly proud that my research on sleep and dreams has been recognized. This acknowledgment validates the years of dedicated work and serves as an inspiration for me to pursue further the practical applications of dream detection in real-world scenarios.”
| Asian Society of Sleep Medicine Interim Meeting 2024 |
|---|