Document Type

Poster

Publication Date

2016

Abstract

Optical blood flow sensors (i.e. photoplethysmographic techniques) have recently been utilized in wearable activity trackers. The Fitbit Charge HRTM (FBHR) is one of the widely recognized wearable activity trackers that utilizes Fitbit’s proprietary PurePulse optical heart rate (HR) technology to automatically measure wrist-based HR. Despite its increasing popularity, however, no study to date has addressed the validity of FBHR for measuring HR in free-living conditions. PURPOSE: The purpose of this study was to examine the validity of FBHR for measuring HR using a chest strap Polar HR monitor (PHR) as a reference measure in free-living conditions. METHODS: Ten healthy college students (8 males; mean age = 26.5 ±5.4 years; mean body mass index (BMI) = 24.5 ±3.23 kg·m2) participated in the study. The participants were asked to perform normal daily activities for 8 hours in a day while wearing the PHR (model RS400) on their chest and two FBHRs on their dominant and non-dominant wrists, respectively. HR was recorded every minute and the minute-by-minute HR data from each monitor were synchronized by time of day. Pearson correlation was used to examine the linearity of average beats-per-minute (bpm) estimated from FBHRs with respect to the PHR. Mean differences in average bpm between the monitors were examined by a general linear model for repeated measures. Lastly, mean absolute percentage error (MAPE) of minute-by-minute bpm estimated from the FBHRs were calculated against the PHR. RESULTS: Average HRs (mean ±SD) for PHR, FBHR non-dominant, and FBHR dominant were 75.6 ±18.5 bpm, 72.8 ±16.7 bpm, and 73.9 ±17.06 bpm, respectively. Pearson correlation coefficients (r) between the PHR and FBHR non-dominant and dominant were r=.805 and r=.793, respectively. MAPE were 9.17 ±10.9% for FBHR non-dominant and 9.71 ± 12.4% for FBHR HR dominant. ANOVA and post-hoc analyses with Bonferroni revealed significant differences in estimating HR from FBHR non-dominant wrist (p=.001) and FBHR dominant wrist (p=.001) compared to PHR monitor. CONCLUSION: The results indicated that the wrist-oriented Fitbit Charge HRTM device does not provide an accurate measurement of HR during free-living condition in this study. However, further research is needed to validate these monitors with a larger sample with different population groups.

Optical blood flow sensors (i.e. photoplethysmographic techniques) have recently been utilized in wearable activitytrackers. The Fitbit Charge HRTM (FBHR) is one of the widely recognized wearable activity trackers that utilizesFitbit’sproprietary PurePulse optical heart rate (HR) technology to automatically measure wrist-based HR. Despiteits increasing popularity, however, no study to date has addressed the validity of FBHR for measuring HR in free-living conditions. PURPOSE: The purpose of this study was to examine the validity of FBHR for measuring HRusing a chest strap Polar HR monitor (PHR) as a reference measure in free-living conditions. METHODS: Tenhealthy college students (8 males; mean age = 26.5 ±5.4 years; mean body mass index (BMI) = 24.5 ±3.23kg·m2) participated in the study. The participants were asked to perform normal daily activities for 8 hours in a daywhile wearing the PHR (model RS400) on their chest and two FBHRs on their dominant and non-dominant wrists,respectively. HR was recorded every minute and the minute-by-minute HR data from each monitor weresynchronized by time of day. Pearson correlation was used to examine the linearity of average beats-per-minute(bpm) estimated from FBHRs with respect to the PHR. Mean differences in average bpm between the monitorswere examined by a general linear model for repeated measures. Lastly, mean absolute percentage error (MAPE)of minute-by-minute bpm estimated from the FBHRs were calculated against the PHR. RESULTS: Average HRs(mean ±SD) for PHR, FBHR non-dominant, and FBHR dominant were 75.6 ±18.5 bpm, 72.8 ±16.7 bpm, and73.9 ±17.06 bpm, respectively. Pearson correlation coefficients (r) between the PHR and FBHR non-dominantand dominant were r=.805 and r=.793, respectively. MAPE were 9.17 ±10.9% for FBHR non-dominant and 9.71 ±
12.4% for FBHR HR dominant. ANOVA and post-hoc analyses with Bonferroni revealed significant differences inestimating HR from FBHR non-dominant wrist (p=.001) and FBHR dominant wrist (p=.001) compared to PHRmonitor. CONCLUSION: The results indicated that the wrist-oriented Fitbit Charge HRTM device does not providean accurate measurement of HR during free-living condition in this study. However, further research is needed tovalidate these monitors with a larger sample with different population groups.

Download

Share

COinS