Preliminary assessment of the Samsung Galaxy Watch 5 accuracy for the monitoring of heart rate and heart rate variability parameters

This dataset was created in the context of the EU project Tolife, which aims at validating an artificial intelligence solution to process daily life patient data captured by non-obtrusive sensors to enable personalized treatment, assessment of health outcomes and improved quality of life in chronic obstructive pulonary disease patients.

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. To this aim, we have created a dataset to evaluate 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 condition. 

The dataset consiststs of photopletismographic (PPG) signal acquired using the SGW5 and Shimmer GSR+, as well as ECG signal acquired using the Shimmer3 ECG, from 20 healthy volunteers performing 1min of resting-state with the arm resting on a table and 1min of walk respectively. 

Both PPG and ECG data have been synchronized using their respective timestamps. The data is given without any preprocessing.Shimmer PPG and ECG data is sampled at 404Hz. SGW5 PPG data is sampled at 25Hz. 

When using this dataset, please cite Rho, Gianluca, et al. "Preliminary Assessment of the Samsung Galaxy Watch 5 Accuracy for the Monitoring of Heart Rate and Heart Rate Variability Parameters." Mediterranean Conference on Medical and Biological Engineering and Computing. Cham: Springer Nature Switzerland, 2023.