Objective: To evaluate the clinical performance of the OPPO Watch (OW) Sleep Analyzer (OWSA) on OSA screening with polysomnography reference.
Methods: We recruited 350 participants using OWSA and PSG simultaneously in a sleep laboratory. The respiratory event index (REI) derived from OWSA and the apnea-hypopnea index (AHI) provided by PSG were compared. SHapley Additive exPlanation (SHAP) values were calculated to explain the model of OWSA.
Results: The OWSA-REI (26.5± 18.5 events/h) correlated well with PSG-AHI (33.2± 25.7 events/h; r = 0.91, p < 0.001), with an intraclass correlation coefficient (ICC) of 0.83. Using a threshold of AHI ≥ 15 events/h, the sensitivity, specificity, accuracy, and area under the curve (AUC) were 86.1%, 86.7%, 86.3%, and 0.94, respectively. Bland-Altman analysis showed that OWSA-REI and PSG-AHI were in good agreement (Mean Difference: − 6.7, 95% CI:16.0 to − 29.3 events/h). In addition, the effectiveness of the models in OWSA were also explained by visualizing SHAP values.
Conclusion: The OWSA demonstrated a reasonable performance for OSA screening in the clinical setting. In light of this, it is possible for smartwatches to become a complementary tool to PSG, which is particularly useful for larger-scale preliminary screenings.

Plain Language Summary: OPPO Watch Sleep Analyzer (OWSA), an emerging sleep-tracking method based on a wearable device, uses a machine learning model to analyze physiological signals including snoring recordings, and other basic anthropometric information to estimate REI. The study evaluated the OSA screening performance of OWSA with PSG-AHI and interpreted the results of the pre-trained machine learning model.
OWSA demonstrated consistent clinical diagnostic performance for OSA. The interpretive machine learning models used in OWSA highlighted the impact of multi-modal data on estimation results. Therefore, these types of models are likely to be more widely accepted and promoted in clinical practice.

Keywords: obstructive sleep apnea, photoplethysmography, polysomnography, smartwatch