Event registration banner
 

Multilevel Stress Classification from ECG Signals

October 20, 2016
9:20 am - 9:40 am
GRB 310 B-C

Track: Data Science
Type: Presentation
Level: Advanced

This study focuses on ECG monitoring for stress detection, which can now be performed with minimally invasive wearable patches and sensors. We have developed an efficient and robust mechanism for accurate stress identification based on feature mining, multiscale entropy, feature selection, and machine learning capable of predicting three stress levels: low, medium and high from ECG signals alone.

Speaker(s)

Isabelle Bichindaritz, PhD, State University of New York at Oswego