Validation of the Samsung Smartwatch for Sleep–Wake Determination and Sleep Stage Estimation

Objectives: Galaxy Watch 3 (GW3) is a commercially available smartwatch equipped with sleep-tracking function capable of collecting longitudinal sleep data in real-world environment. We aimed to investigate the validity GW3 for estimating stages compared reference from polysomnography (PSG).Methods: Thirty-two healthy adults (mean age 37.8, male 87.5%) were recruited wear concurrently in-laboratory overnight PSG recording. Sleep parameters, including total time (TST) and duration each stage (light, deep, rapid eye movement [REM] sleep), calculated both PSG. parameters using intraclass correlation coefficients (ICCs) Bland–Altman plots. The epoch-by-epoch classification performance was evaluated determine sensitivity, specificity, accuracy, kappa values, confusion matrices.Results: plots showed moderate agreement between TST (ICC=0.640), light (ICC=0.518), deep (ICC=0.639), whereas REM not reliably estimated GW3. overestimated by mean 9.5 min. sensitivity detection 0.954; however, specificity 0.524. estimation 0.695 sleep, 0.612 0.598 sleep. overall accuracy distinguishing four-stage epochs 0.651.Conclusions: demonstrated high but wake determination results, which comparable previously reported results other consumer wearable devices.