| Start Date: | 6/17/2022 | Start Time: | 2:00 PM |
| End Date: | 6/17/2022 | End Time: | 4:00 PM |
|
Event Description
BIOMED Master's Thesis Defense
Title:
Implementation of a Home-Use Virtual Environment Brain-Computer Interface (BCI) for People with ALS Using Different Facial Stimuli
Speaker:
Emma Dryden, Master’s Candidate
School of Biomedical Engineering, Science and Health Systems
Drexel University
Advisor:
Hasan Ayaz, PhD
Associate Professor
School of Biomedical Engineering, Science and Health Systems
Drexel University
Details:
Amyotrophic Lateral Sclerosis (ALS) is a neurodegenerative disease that causes progressive loss of voluntary movement, including the ability to speak. As the disease rapidly progresses, there is a need for augmented and alternative communication that removes physical barriers required for engagement. Brain-Computer Interfaces (BCI) are a technology that allow for communication using only brain signals in response to visual stimuli. This project aimed to design an accessible and practical BCI for continuous, at-home use, and improve the brain response with familiar facial stimuli to increase BCI usability. This was sought out with the overall goal to maximize ease of use for individuals with ALS and their caregivers so that BCI technologies can make a useful impact and improve the everyday lives of people with ALS.
First, a new miniaturized and low-cost P300 BCI system was designed using wireless and battery-operated EEG neuroimaging, modular software, and custom stimuli presentation. A 3-dimensional virtual environment platform was integrated as a navigational control output mechanism that executes commands received from the BCI to move a users’ avatar in a virtual maze. These together form the virtual environment BCI (veBCI). Finally, for BCI stimuli presentation, we developed a custom approach of integrating different human face images with the aim of increasing user engagement and elicited brain activity, and therefore improve BCI accuracy. We tested the impact and effectiveness of our approach via a verification BCI study in which we utilized 6 different stimuli conditions: male celebrity face, female celebrity face, male stranger face, female stranger face, a family member/close friend of the participant, and blank/no face as control. A total of 16 volunteers (average age 56.5 and eight diagnosed with ALS) used the BCI system with all 6 face conditions in ecologically valid settings. Results indicate that this new miniaturized BCI is usable for both people with ALS and healthy controls. Significant differences did not exist between the P300 response in ALS and healthy controls. However, differences between male and female participants demonstrated the potential role of personalization to optimize the visually evoked potential response and further increase BCI accuracy.
In conclusion, a home-use P300 BCI system was developed with a new generation of miniaturized, low-cost and mobile neuroimaging running on an embedded computational platform. BCI stimuli with human face inclusion could help personalization of the BCI. Furthermore, virtual environments provide unique new ways to implement BCI output beyond typical spelling tasks and communication. Future work could focus on developing dedicated online signal processing to eliminate noise and improve accuracy during home use. |
|
Audience:
Undergraduate StudentsGraduate StudentsFacultyStaff |
|
