Abstract: The devastating impact of stress on a wide variety of health conditions warrants the creation of new tools for continuous stress monitoring in ambulatory settings. In this poster, we assess the quality of heart rate variability and electrodermal activity data obtained from wrist-worn sensors in the context of stress monitoring. We discuss novel algorithms for improving data quality and increasing the amount of information that can be extracted from these biosignals.
Learning Objective 1: After viewing this poster, the learner should be better able to:
1) Select an algorithm for inferring stress levels from heart rate variability (HRV) and electrodermal activity data (EDA) collected using wrist-sensors.
2) Select an appropriate algorithm for dealing with noisy HRV and EDA data in the context of ambulatory stress monitoring.
3) Apply this knowledge to their own practice.
Maciej Kos (Presenter)
Christine Gordon, Northeastern University
Xuan Li, Northeastern University
Iman Khaghani-Far, Northeastern University
Misha Pavel, Northeastern University
Holly Jimison, Northeastern University