0775 SLEEP STAGE ESTIMATION USING A NON-CONTACT BIOSENSOR

Abstract

Introduction: Polysomnography is a gold standard sleep evaluation mainly conducted in sleep labs. Repetitive evaluation of sleep quality at home could be useful in the management of insomnia especially when there is discrepancy between subjective and objective sleep. The aim of the study was to evaluate the characteristic of a non-contact sensor in identifying sleep stages. Methods: Fifteen patients who underwent diagnostic polysomnography (PSG) at the university sleep laboratory were recruited. Non-contact biosensor measuring heart rate, respiration and body movement was laid under a bed mattress. Sleep stages were scores based on a AAMS standard procedure for PSG, and sleep stage analysis for the non-contact device was made using a previously reported method (Harada et al., 2016). Pattern of hypnogram obtained from standard PSG and non-contact device was compared. The study was approved by the institutional review board and written informed consent was obtained. Results: Recordings were successfully completed in all cases. On average, non-contact device obtained heart rate and respiration data in 89.2% and 70.6% of the recording time respectively. Non-contact device mostly identified the sleep architecture trend, but non-contact device tend to overestimate slow wave sleep, especially in patients with repetitive arousals or awakenings. Conclusion: Non-contact sensor equipped under the bed mattress was able to identify clinically useful sleep quality data. Some modification in the detection of respiration signal and the analysis argorhythm idetifying brief awakenings may be necesarry.