I took EC601: Product Design in Electrical and Computer Engineering during my first semester of graduate study at Boston University. Throughout the semester my partner and I worked on developing a non-invasive, EEG-based brain computer interface (BCI) for smart home applications specifically aimed at patients with ALS. We used hardware from OpenBCI to acquire EEG data, and fed it into a compact neural network called EEGNet. We trained EEGNet on a dataset of simple motor tasks, and implemented the pre-trained model for real-time classification of incoming EEG data. We created a demo video, and all of our scripts are available on GitHub.

While our final prototype was not able to reliably classify the motor tasks from our live data stream, working on the project was a great introduction to working with AI models, as well as working with BCIs in general. My experiences and excitement towards this project motivated me to enroll in an introductory machine learning course for the next semester, and my partner and I are continuing to work on the project in our spare time.