In the last years, commercial smartwatches have gained popularity as non-invasive and wearable devices to be exploited for the monitoring of the cardiovascular system in daily-life settings. However, their reliability is still unclear. In this preliminary study, we evaluated the accuracy of heart rate (HR) and HR variability (HRV) estimates obtained from the Samsung Galaxy Watch 5 (SGW5) compared to a common research-grade ECG sensor, i.e., the Shimmer3 ECG unit (ShimECG), during both a resting and walking conditions. For each condition, we compared HRV features of SGW5 and ShimECG extracted in time, frequency, and non-linear domains through correlation and Bland-Altman analysis. Additionally, we compared SGW5 performance with those obtained from a research-grade PPG sensor. Our results revealed an unbiased and high-quality estimate of mean HR obtained from the SGW5. Moreover, at rest, other relevant HRV features showed a significant correlation between the SGW5 and ShimECG. Conversely, during the walking condition, we found poor performances for both PPG devices for most of the HRV features. Such preliminary results confirm the reliability of SGW5 to estimate mean HR. However, the reliability of SGW5- derived PRV to extract sympathovagal correlates is still an open question and deserves further investigation.
Preliminary Assessment of the Samsung Galaxy Watch 5 Accuracy for the Monitoring of Heart Rate and Heart Rate Variability Parameters
Rho G.;Di Rienzo F.;Marinai C.;Giannetti F.;Arcarisi L.;Bufano P.;Zanoletti M.;Righetti F.;Vallati C.;Laurino M.;Carbonaro N.;Tognetti A.;Greco A.
2024-01-01
Abstract
In the last years, commercial smartwatches have gained popularity as non-invasive and wearable devices to be exploited for the monitoring of the cardiovascular system in daily-life settings. However, their reliability is still unclear. In this preliminary study, we evaluated the accuracy of heart rate (HR) and HR variability (HRV) estimates obtained from the Samsung Galaxy Watch 5 (SGW5) compared to a common research-grade ECG sensor, i.e., the Shimmer3 ECG unit (ShimECG), during both a resting and walking conditions. For each condition, we compared HRV features of SGW5 and ShimECG extracted in time, frequency, and non-linear domains through correlation and Bland-Altman analysis. Additionally, we compared SGW5 performance with those obtained from a research-grade PPG sensor. Our results revealed an unbiased and high-quality estimate of mean HR obtained from the SGW5. Moreover, at rest, other relevant HRV features showed a significant correlation between the SGW5 and ShimECG. Conversely, during the walking condition, we found poor performances for both PPG devices for most of the HRV features. Such preliminary results confirm the reliability of SGW5 to estimate mean HR. However, the reliability of SGW5- derived PRV to extract sympathovagal correlates is still an open question and deserves further investigation.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
