Potential participants with suspected OSA will be identified from the schedule of the TMU sleep labs. Those subjects who satisfy the study conclusion and exclusion criteria will be approached and invited to participate in the study.

BSP(Belun Technology Company Limited) is a novel wearable system using a neural network algorithm that consists of a patented wearable device (Belun Ring), a charging cradle, and cloud-based software. The Belun Ring, an FDA-cleared pulse oximeter, acquires signals from the radialis indicis artery of the proximal index finger. The Belun Ring total sleep time can be derived from features extracted from accelerometer, SpO2, and PPG signals, whereas Belun Ring AHI can be derived from Ring-TST and features extracted from HRV and SpO2 changes. The BSP proprietary OSA detection algorithm was using neural network. BSP performance can be improved by training the algorithm through continual data collection. To our knowledge, BSP is the only validated sleep platform using a medical-grade wearable pulse oximeter, actigraphy, and artificial intelligence algorithm that has the capability to identify sleep stages, detect OSA events, and monitor autonomic nervous system activity changes during sleep.

To specifically assess the accuracy of BSP bAHI in predicting OSA by comparing to the concurrent in-lab PSG-AHI and to determine the accuracy of BSP sleep stage parameters by comparing to the concurrent PSG.

BSP bAHI, BSP time with SpO2 < 90%, and BSP-SpO2 nadir will be extracted from BSP and compared to PSGAHI (4% hypopnea criteria), PSG time with SpO2 < 90% (PSG-T90), PSG-SpO2 nadir extracted from the concurrent PSG.

BSP sleep stage parameters (total sleep time [TST], wake time, REM time, and NREM time) will be extracted from BSP and compared to the same parameters of the concurrent PSG. Epoch-by-epoch comparison will be performed.

Duration of BSP use and technically valid recording time will be extracted from BSP for assessment feasibility of BSP testing in hospitalized patients.