As part of virtual micro courses that will be taking place throughout the semester, Carnegie Mellon faculty will be giving virtual lectures open to the CMU-Q community. Mayank Goel, Assistant Teaching Professor at the Institute for Software Research, School of Computer Science at Carnegie Mellon University will be speaking on, “Recognized User’s Actions and Physiology for Health”.
In the last decade, many researchers have explored the timely detection of mental health changes using sensors in the user’s environment. This is mostly accomplished using wearables; especially in solutions that are deployed in the real world. Wearables are sensor-laden, extremely personal, almost always on the user. The challenge is that the sensors on most wearables only provide a coarse estimate of a user’s actions and behavior. The ideal sensing system would be inexpensive, use existing sensors, be plug-and-play, and able to be deployed with no training or calibration. The talk will present our efforts towards building a general-purpose and flexible activity and physiology inference system that relies on an ecosystem of smart devices while keeping deployment, training, and user burden to a minimum.