Year
2026
Season
Spring
Paper Type
Master's Thesis
College
Brooks College of Health
Degree Name
Master of Science in Health: Kinesiology (MSH)
Department
Clinical & Applied Movement Sciences
NACO controlled Corporate Body
University of North Florida. Department of Clinical & Applied Movement Sciences
Committee Chairperson
Dr. Lindsay P. Toth
Second Advisor
Dr. James Churilla
Rights Statement
http://rightsstatements.org/vocab/InC/1.0/
Third Advisor
Dr. Tammie Johnson
Fourth Advisor
Dr. Annie Mulholland
Department Chair
Dr. Joel Beam
College Dean
Dr. Mei Zhao
Abstract
BACKGROUND: Wrist-worn wearable activity monitors have presented mixed evidence regarding their heart rate (HR) accuracy across the skin tone spectrum and exercise intensity spectrum. Additionally, controversial subjective measures were used. Furthermore, not all the studies included exercise or tested limited modalities. Therefore, the aims of this study are to 1) assess the HR error of several wearables during rest, aerobic exercise, and exergaming across the objectively measured skin tone spectrum; 2) assess the HR accuracy of several wearables during rest, aerobic exercise, and exergaming compared to a criterion measure for HR; and 3) to compare HR error between the criterion measurement and predicted heart rate when skin tone is measured objectively vs subjectively.
METHODS: Healthy adults (N=44) were asked to wear four wearable activity monitors and a Polar chest strap to measure HR during rest, treadmill walking, jogging, running, arm ergometry, stair climbing, and exercise gaming. Descriptive statistics were calculated to find the mean bias and mean absolute percent error of HR. Additionally, linear mixed models were used to assess the interactions on skin tone, activity, and monitor on HR accuracy. Modified Bland-Altman plots were used to visually assess the systemic bias.
RESULTS: Empatica showed the greatest overall error when compared to Garmin and Fitbit (MAPE = 12.15 ± 17.10%; 9.18 ± 16.65%; 8.55 ± 13.08%, respectively) There was no significant main effect on the three-way interaction between monitor, activity, and objective skin tone, F(12, 854.04) = 0.40, p = 0.965. There was a significant interaction between monitor and activity, F(12, 854.10) = 2.57, p = 0.002.
CONCLUSION: Our findings suggest that there is no effect on HR accuracy based on skin tone. However, there is a significant effect of activity type on HR accuracy of wrist-worn wearables.
Suggested Citation
Rollins, Isabella G., "Heart rate accuracy of wearable activity monitors across the skin tone and exercise intensity spectrum" (2026). UNF Graduate Theses and Dissertations. 1407.
https://digitalcommons.unf.edu/etd/1407
Accessibility Statement
This item was created or digitized before April 24, 2027, or is a reproduction of legacy material created before that date. It is preserved in its original, unmodified state specifically for research, reference, or historical recordkeeping. In accordance with the ADA Title II Final Rule, the Library provides accessible versions of archival materials by request. If you are experiencing difficulty accessing the information on the site due to a disability, please submit a request through the following form for assistance.