Abstract

In this talk, I will introduce our research on building a robust ear-worn system that strategically integrates neurophysiological sensors and computing capabilities around the ears for ambulatory seizure monitoring. The development of such a system presents significant challenges, requiring rethinking of hardware and software frameworks, advanced ML algorithms, and strong interdisciplinary efforts. I will present our approaches to address these gaps and develop a practical ear-worn system for seizure detection and classification in real-world settings. Our work focuses on designing robust signal extraction algorithms to capture brain activity at the ear, developing techniques to extend battery lifetime of the earable, identifying seizure-related biomarkers from the ears’ signals, and validating the system through hospital deployments at the University of Texas Southwestern Medical Center and UMass Chan Medical School. I will also share insights gained from deploying these systems in real-world settings and discuss future research directions.

Bio: VP Nguyen directs the Wireless and Sensor Systems Laboratory (WSSL) at the Manning College of Information and Computer Sciences, University of Massachusetts Amherst, and is also affiliated with the Institute for Applied Life Sciences. He is the recipient of ACM SIGMOBILE Research Highlights (2017, 2020, 2022), Sony Faculty Innovation Award, and CACM Research Highlights (2020, 2021). His broad research interests span mobile and wearable computing, wireless networking, and embedded systems, with a particular focus on the Internet of Medical Things (IoMT) for disease monitoring, intervention, and prevention. His lab research aims to improve healthcare accessibility by developing mobile and wearable health technologies that are portable, reliable, comfortable to use, socially acceptable, and low-cost. Their work is highly interdisciplinary, advancing sensor and sensing system development, algorithms, data analytics, signal processing, hardware, and firmware optimization to invent a new class of medical devices capable of continuously and unobtrusively collecting clinically meaningful health data in non-clinical settings.