Project

Our team is developing novel language processing and predictive algorithms, as well as responsive wearable systems that look and feel like clothing. The results will transform the standard of care for groups such as the aging population, which is expected to grow from 524 million worldwide in 2010 to 1.5 billion in 2050.

Year 1

Stress is ubiquitous. It's also an important factor in the development and progression of multiple diseases, and it varies substantially in the extent to which people encounter stress and the way they respond to it. Current methods for detecting and managing stress in people’s natural environments are inadequate and methods for intervening in real-time are largely lacking. For these reasons and more, we selected the detection and management of stress and anxiety as the initial target for a proof-of-concept framework.

We are conducting a study in a controlled, lab-based setting with University of Minnesota students, with the goal of collecting the data needed to develop conversational agent capabilities. In this study, the conversational agent interacts with the user twice a day: once in the morning to gather information on potential stressors anticipated during the day; and once in the evening to gather information on stressors that actually occurred during the day.

The information on the daily stressors—including number, timing, severity, and recurrence—is used to help interpret noisy physical and electrodermal activity and heart rate sensor signals. This data helps us maximize the accuracy and reliability of detecting physiological response to stressors.

At the same time, we will conduct another lab-based feasibility study to test the ability of the compression garment to reduce the subjective and physiological response to a stressor presented in a controlled laboratory experiment.

Year 2

During Year 2, we will use the framework developed in Year 1 in a natural environment. We will conduct a feasibility study in which 10-12 participants recruited from the elderly population in the broader University community. These faculty, staff, and alumni will represent an older population than the student participants in Year I. They will be asked to wear the Empatica E4 device and the compression garment for two-day observation periods in their natural environments. They will also interact with the system by using an Amazon Echo in their homes.

The objective is to test the integration of the components of the proposed framework and to evaluate its ability to serve as a foundation, and provide preliminary data, for larger subsequent projects. This study will also provide pilot data regarding the effectiveness of compression as a stress reduction intervention in the natural environment. The exact details of the study design will be informed by the results of Phase I studies.

This study will be used to determine the characteristics of the entire framework when used outside the laboratory (e.g., energy consumption, network and Bluetooth connectivity, physical robustness). It will also bring to light any potential issues and limitations associated with everyday use. This information will guide continued development in larger, subsequent studies that focus on stressors and other social and behavioral determinants of health.