MIT EECS Lal Undergraduate Research and Innovation Scholar
Machine Learning for Embedded Analytics: Electromyography for Bruxism Detection
Electrical Engineering and Computer Science
Anantha P. Chandrakasan
Low-Power Biomedical Sensors and Machine Learning
Stress-related disorders like migraines and bruxism have symptoms that can cause consistent pain inhibiting productivity and quality of life. I will be working to develop efficient machine learning algorithms that can be programmed onto a biopotential sensor that are capable of providing diagnostic information on electromyographic signals. We desire understanding of how stress-related diseases manifest themselves in the electric signals of our facial muscles. By analyzing how the signals propagate through space and time there is potential to understand these disorders in a computationally relevant way. Once the electrical properties of the muscles are understood it could be possible to stimulate the muscles in a closed-loop fashion to provide relaxation and relief to the patients.
My name is Andrew Mullen and I am a senior studying EECS at MIT. I’ll be developing machine learning algorithms for low-power sensors detecting electromyographic signals. I have been doing biomedical research at MIT since my sophomore year first at Biomechatronics in the Media Lab and then neural engineering work at the Picower Institute. My intellectual interests lie at the intersection of electronics and biology.