Sriyan Madugula
Pronouns: he/him
Research Mentor(s): Jane Huggins
Research Mentor School/College/Department: Physical Medicine and Rehabilitation / Medicine
Program:
Authors: Sriyan Madugula, Jane E. Huggins
Session: Session 3: 11:00 am – 11: 50 am
Poster: 24
Abstract
Brain-computer interfaces (BCIs) enable direct communication between the brain and external devices, and a particular application of them is being developed by the University of Michigan Direct Brain Interface (UM-DBI) Laboratory to help people with disabilities handle technology. The BCI interprets brain activity in response to flashes of keys and shows whether a flash of a particular group of keys contains the key that the user wants to type. As the BCIs are still under development, a study was undertaken to test the long-term stability of the calibration data used to set up the BCI for a particular user. Hundreds of data files were indexed in an Excel spreadsheet. The spreadsheet detailed aspects of BCI calibration and testing data from 2008 to 2023, such as the sample rate, electrodes, the amount of data available, and the type of stimuli used. Files containing usable data were noted. This information can be used to package the data for future analysis through the data packaging program. For our preliminary analysis, we will plot brain activity through the P300 classifier, which is a specific BCI UM-DBI uses. We plan on plotting brain activity for training data spaced out on 2-5 year intervals to analyze their differences. We expect to finish this portion of the analysis by the end of the term, but a more thorough analysis of the long-term stability of the calibration data remains to be done
