Stress in the Age of Technology: Evaluating the Validity and Accessibility of Cardiac Data from Industry-Leading Smartwatches

Author:
Moore, Aaron, Systems Engineering - School of Engineering and Applied Science, University of Virginia
Advisor:
Riggs, Sara, EN-Eng Sys and Environment, University of Virginia
Abstract:

Objective: To test the validity and data accessibility of the health detection features on two leading smartwatches in the industry, the Apple Watch Series 6 and Fitbit Sense, and to determine if these devices can be used to accurately measure stress responses in the body.
Background: Stress causes the release of hormones in the human body that increase heart rate, blood pressure, and energy supplies. Over time, the constant release of stress hormones can lead to many medical, psychological, and behavioral health problems. Health problems related to or caused by stress have become increasingly prevalent in the modern world. Access to health information on mobile devices have allowed users to more easily keep track of their fitness, but there is still much to be learned when it comes to stress detection. The ability to detect stress responses and access the measured data needs to be further examined.
Method: Sixteen participants performed modules of a multi-tasking program while smartwatches measured cardiac information. Multi-tasking modules varied in difficulty and included low, medium, and high workload conditions.
Results: Workload has an effect on the amount of stress responses produced by the participants, as the testing conditions showed an increase in stress responses compared to the baseline. The presence of detectable stress responses aligned with the participant's perception of how hard they were working and how stressed they were feeling. The Apple Watch Series 6 was unable to detect significant differences in stress responses as compared to the baseline in this study, and did not align with the measurements of the official ECG device used.
Conclusion: The accessibility of health-related data for in depth analysis on current smartwatches is very limited, and is highly susceptible to variability when measurements are being taken. Data collection limitations in current smartwatches make it difficult to capture peak stress responses during varying workload conditions, and likely was the cause of insignificance. Stress detection in smartwatches must be investigated further.
Application: The findings of this research provide insight on the validity and accessibility of health metrics gathered from current smartwatches and how stress responses can be evaluated using these devices.

Degree:
MS (Master of Science)
Keywords:
smartwatches, stress detection, heart rate, heart rate variability, workload
Language:
English
Rights:
All rights reserved (no additional license for public reuse)
Issued Date:
2021/12/12