Marvin Jirapongsuwan
Research Mentor(s): Jane Huggins
Mentor Department: Physical Medicine and Rehabilitation
Authors: Marvin Jirapongsuwan, Jane Huggins
Session: Session 3 (11:00am – 11:50am)
Presentation Type: Poster 48
Abstract
Brain-computer interfaces (BCIs) have transformative potential, especially for individuals with severe disabilities. However, challenges such as accuracy, scalability, and latency hinder their widespread use. At the University of Michigan Direct Brain Interface (UMDBI) Lab, this project focuses on improving the existing codebase of custom BCI modules designed to enhance communication and cognitive assessment, using the BCI2000 framework. One module that the project works on, the Augmentative and Alternative Communication (AAC)-BCI module, empowers communication by translating users’ brain activity into key selections on a virtual keyboard. Another, the UMDBIStimulusPresentationTask presents paired images, with the results used for calibration and further exploration of decision-making in other custom modules developed by the lab. Specifically, this project focuses on improvements of the real-time BCI system by developing robust error checks, resolving core system crashes, and implementing new functionality. In the AAC-BCI module, the project mainly implements error handling, and bug fixes. Whilst in the UMDBIStimulusPresentationTask, the project expands new functionality to provide operators with greater flexibility in proctoring patients. Through the project’s contribution to the codebase, the lab now has new functionality previously unavailable.
