수면무호흡증 예측시 연속적인 스마트워치 수면추적의 유용성
Usefulness of Continuous Sleep Tracking by Smartwatch in Predicting Sleep Apnea
3
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Background and Objectives The purpose of this study was to determine whether smartwatch (SW) data measured over several nights at home could effectively predict obstructivesleep apnea (OSA). Subjects and Method Patients suspected of having OSA and scheduled for polysomnography (PSG) were asked to wear a SW at home for 7 nights before undergoing PSG. After obtaining the SW data, it was compared and analyzed with the PSG. Results A total 59 people including 42 OSA patients and 17 normal control were included inthe study. The sensitivity and specificity for predicting apnea-hypopnea index (AHI) ≥5/h using the SW lowest oxygen saturation of 85% were 85.7% and 70.6%. The area under curve(AUC) for the receiver operating characteristic (ROC) curve was 0.901 (p<0.001). The sensitivity and specificity for predicting AHI ≥15/h using the SW lowest oxygen saturation of 82%were 80.6% and 67.9%. The AUC for the ROC curve was 0.866 (p<0.001). Conclusion This study showed that the OSA can be predicted fairly accurately using dataobtained from wearing a SW for several nights in an environment similar to daily life ratherthan a hospital laboratory.
Abstract
Background and Objectives The purpose of this study was to determine whether smartwatch (SW) data measured over several nights at home could effectively predict obstructivesleep apnea (OSA). Subjects and Method Patients suspected of having OSA and scheduled for polysomnography (PSG) were asked to wear a SW at home for 7 nights before undergoing PSG. After obtaining the SW data, it was compared and analyzed with the PSG. Results A total 59 people including 42 OSA patients and 17 normal control were included inthe study. The sensitivity and specificity for predicting apnea-hypopnea index (AHI) ≥5/h using the SW lowest oxygen saturation of 85% were 85.7% and 70.6%. The area under curve(AUC) for the receiver operating characteristic (ROC) curve was 0.901 (p<0.001). The sensitivity and specificity for predicting AHI ≥15/h using the SW lowest oxygen saturation of 82%were 80.6% and 67.9%. The AUC for the ROC curve was 0.866 (p<0.001). Conclusion This study showed that the OSA can be predicted fairly accurately using dataobtained from wearing a SW for several nights in an environment similar to daily life ratherthan a hospital laboratory.