A research team from the University of Granada has studied how wearable EEG headbands can be used with artificial intelligence to detect human emotions. This research, led by Francisco M. García-Moreno—member of the MYDASS Research Group and R&D Project On the Edge Project—applies deep learning to classify emotional states in terms of arousal and valence using brainwave signals captured by low-cost devices.
Understanding Emotions Through Brain Signals
In this work, the researchers evaluated how the EEG signals collected by wearable headbands, such as the Muse device, can be used to detect emotional states. Unlike traditional systems that rely on high-density caps with dozens of electrodes, this study explores the limits and possibilities of using just a few channels. “The idea was to determine which brainwave bands contribute the most to identifying emotional responses,” says García-Moreno.
The study, published in the journal Computers in Biology and Medicine, achieved promising classification results using a minimal number of electrodes and only selected frequency bands. The findings contribute to the field of affective computing by opening new possibilities for emotion-aware systems in realistic and portable settings.
Applications in Healthcare and Human–Machine Interaction
Potential applications include mental health monitoring, adaptive learning platforms, and emotionally responsive user interfaces. “We’re interested in building bridges between neuroscience, wearable technology, and artificial intelligence,” says García-Moreno. “This line of work could pave the way for future developments in digital mental health tools and emotion-aware devices.”
Open Access Publication and Institutional Support
The full study is available as an open access article through Elsevier’s ScienceDirect platform:
🔗 https://doi.org/10.1016/j.compbiomed.2024.109463
📰 Spanish press coverage by Canal UGR: Read the news article.
This project was supported by the R&D Project On the Edge Project,CITIC-UGR and developed within the framework of the MYDASS research group at the University of Granada.
