Using Wearable Sensors to Predict Pain States

Friday, February 8, 2019 - 1:25pm - 2:25pm
Lind 305
Kyle Srivastava (Boston Scientific)
In all healthcare environments, providers are frequently asking patients how much pain they are in, whether to determine the effectiveness of a therapy or the urgency of a problem. Traditionally, this is measured by asking someone how much pain they are in on a scale of 0 to 10. However, it can be difficult to account for variability across patients as well as other factors within a given patient. Discovering an objective way to measure pain would eliminate the subjectivity associated with this loaded question. In this work, we recorded physiological data using wearable devices in chronic pain patients undergoing different therapies. We then modeled pain in several ways based on those data. While we were able to explain a significant portion of the variance observed in pain scores, there are also challenges that exist in the way of our ultimate goal.

Kyle has worked with Boston Scientific in Arden Hills for 2.5 years as part of the Corporate Research group. In this group, he explores new technology and algorithms as they may relate to all parts of the company. He has a focus on data analysis and machine learning and has worked on projects related to physiological data, electronic health record data, medical imaging data, manufacturing data, and others. Prior to joining the Boston Scientific team, Kyle did his undergraduate work in Bioengineering and Economics at the University of Pennsylvania. He subsequently studied vocal motor control as part of his PhD in Biomedical Engineering with a focus in neural engineering at Georgia Tech and Emory University. In addition to research, he enjoys sharing my love for science with young students, volunteering in the community, and staying active by running and swimming.