Dhruv Dighrasker
Pronouns: He/Him
Research Mentor(s): Jane Huggins
Research Mentor School/College/Department: Physical Medicine and Rehabilitation / Medicine
Program:
Authors: Dhruv Dighrasker, Jane E. Huggins
Session: Session 3: 11:00 am – 11: 50 am
Poster: 26
Abstract
The University of Michigan Direct Brain Interface (UM-DBI) Laboratory is leading the development of a groundbreaking Brain-Computer Interface (BCI) to revolutionize communication for individuals with severe motor and vocal impairments. Leveraging an open-source software platform, [1] BCI2000, UM-DBI’s innovation includes a custom P300 Classifier. The [2] P300 classifier is a signal processing tool that can be used to interpret Electroencephalography (EEG) data by recognizing positive deflections that occur approximately 300 milliseconds after the presentation of a relevant stimulus. The cost and setup time of P300 BCIs, however, are strongly linked to the number of channels, or electrodes, used. UM-DBI’s custom variant is designed to optimize electrode selection, allowing us to artificially limit the maximum number of electrodes that can be used. Over the past few months, several features have been added to the P300 classifier GUI to help with this very goal. The GUI’s details page now displays both the number of channels and a list of each channel used. This information reveals which locations were selected as important even when the number of electrodes that can be used is not restricted.. The details page also outputs several statistical values regarding the scores associated with target (relevant stimuli) and non-target (irrelevant stimuli) that offer insight into the accuracy of the sample data that the classifier is reading. By pioneering a more efficient and adaptive P300 Classifier, UM-DBI Laboratory is broadening the horizons of independence and interaction for the underserved population. [1] https://www.bci2000.org/mediawiki/index.php/Main_Page [2] https://www.bci2000.org/mediawiki/index.php/User_Reference:P300Classifier
