Williams, K;
Jamieson, A;
Chaturvedi, N;
Hughes, A;
Orini, M;
(2023)
Validation of Wearable Derived Heart Rate Variability and Oxygen Saturation from the Garmin's Health Snapshot.
In:
Computing in Cardiology 2023 (CinC 2023).
Computing in Cardiology
Preview |
Text
CinC2023-237.pdf - Accepted Version Download (404kB) | Preview |
Abstract
The new Garmin 'Health Snapshot' feature claims to measure resting heart rate (RHR), heart rate variability (HRV) and oxygen saturation (SpO2) over a two-minute period, providing a 'glimpse of overall cardiovascular status'. This is the first study to investigate the accuracy of the feature in healthy adults (n=27, 63% male, mean ± SD age = 21.9pm 6.7). Slower respiratory rates are known to increase HRV, therefore two respiratory rates (normal and controlled) were incorporated within the protocol. Reference measures for RHR and HRV metrics (RMSSD and SDNN) were derived from an electrocardiogram (ECG), whereas reference SpO2 was determined using a Pulse Oximeter. Health Snapshot accuracy was quantified using Pearson's/Spearman's (cc_{p}/cc_{s}) correlation coefficients, Bland-Altman plots and mean absolute error (MAE). Health Snapshot estimations of RHR produced almost perfect correlation (0.99), MAE < 2% and narrow limits of agreement. Under normal breathing, both HRV metric estimations produced good correlation (cc_{p} > 0.82). SpO2 estimation was relatively poor with ∼16.7% of Garmin estimations < 95%, despite all references ≥ 98%. HRV metric estimations were less accurate during controlled breathing, because wearable-derived HRV was slightly underestimated for lager HRV values.
Archive Staff Only
 |
View Item |
