Brain-Computer Interfaces (BCI)

Abstract

Brain-Computer Interfaces (BCI) attempt to measure neuronal activity in the brain of a computer user and use those measurements to infer the user’s cognitive state. This is a highly interdisciplinary area of study, which overlaps with neuroscience, psychology, and computer science. Applications of BCI include many areas such as interactive media, adaptive user interfaces, accessible user interfaces, usability testing, and human-machine teaming. In this talk, the author will present his work on improving BCI using machine learning. He used a brain activity sensor called functional Near InfraRed Spectroscopy (fNIRS), which uses near-infrared light to measure blood flow in the brain’s cerebral cortex. He developed novel preprocessing and machine learning techniques to analyze fNIRS data and infer user emotion and cognitive workload. He will present the methods and machine learning techniques used in his research. He will also talk about the implications of his research and future directions.

Bio

Danushka Bandara received his Ph.D. in Electrical and Computer Engineering and M.S. in Computer Engineering from Syracuse University in 2018 and 2013, respectively, and a B.S. in Electrical Engineering with honors from the University of Moratuwa in 2009. Before joining Fairfield University, he worked as a Data Scientist at Corning Incorporated. The focus of his Ph.D. research was on the application of machine learning to brain activity data. His research interests include machine learning, human-computer interaction, computer vision, pattern recognition, and signal processing.

Fairfield, Connecticut, United States, Virtual: https://events.vtools.ieee.org/m/289362