Yuan Hang, Plekhanova Tatiana, Walmsley Rosemary, Reynolds Amy C, Maddison Kathleen J, Bucan Maja, Gehrman Philip, Rowlands Alex, Ray David W, Bennett Derrick, McVeigh Joanne, Straker Leon, Eastwood Peter, Kyle Simon D, Doherty Aiden
Nuffield Department of Population Health, University of Oxford, UK.
Big Data Institute, Li Ka Shing Centre for Health Information and Discovery, University of Oxford, UK.
medRxiv. 2023 Jul 8:2023.07.07.23292251. doi: 10.1101/2023.07.07.23292251.
Sleep is essential to life. Accurate measurement and classification of sleep/wake and sleep stages is important in clinical studies for sleep disorder diagnoses and in the interpretation of data from consumer devices for monitoring physical and mental well-being. Existing non-polysomnography sleep classification techniques mainly rely on heuristic methods developed in relatively small cohorts. Thus, we aimed to establish the accuracy of wrist-worn accelerometers for sleep stage classification and subsequently describe the association between sleep duration and efficiency (proportion of total time asleep when in bed) with mortality outcomes.
We developed and validated a self-supervised deep neural network for sleep stage classification using concurrent laboratory-based polysomnography and accelerometry data from three countries (Australia, the UK, and the USA). The model was validated within-cohort using subject-wise five-fold cross-validation for sleep-wake classification and in a three-class setting for sleep stage classification wake, rapid-eye-movement sleep (REM), non-rapid-eye-movement sleep (NREM) and by external validation. We assessed the face validity of our model for population inference by applying the model to the UK Biobank with 100,000 participants, each of whom wore a wristband for up to seven days. The derived sleep parameters were used in a Cox regression model to study the association of sleep duration and sleep efficiency with all-cause mortality.
After exclusion, 1,448 participant nights of data were used to train the sleep classifier. The difference between polysomnography and the model classifications on the external validation was 34.7 minutes (95% limits of agreement (LoA): -37.8 to 107.2 minutes) for total sleep duration, 2.6 minutes for REM duration (95% LoA: -68.4 to 73.4 minutes) and 32.1 minutes (95% LoA: -54.4 to 118.5 minutes) for NREM duration. The derived sleep architecture estimate in the UK Biobank sample showed good face validity. Among 66,214 UK Biobank participants, 1,642 mortality events were observed. Short sleepers (<6 hours) had a higher risk of mortality compared to participants with normal sleep duration (6 to 7.9 hours), regardless of whether they had low sleep efficiency (Hazard ratios (HRs): 1.69; 95% confidence intervals (CIs): 1.28 to 2.24 ) or high sleep efficiency (HRs: 1.42; 95% CIs: 1.14 to 1.77).
Deep-learning-based sleep classification using accelerometers has a fair to moderate agreement with polysomnography. Our findings suggest that having short overnight sleep confers mortality risk irrespective of sleep continuity.
睡眠对生命至关重要。准确测量和分类睡眠/清醒状态及睡眠阶段，对于睡眠障碍诊断的临床研究以及解读用于监测身心健康的消费设备数据而言非常重要。现有的非多导睡眠图睡眠分类技术主要依赖于在相对较小队列中开发的启发式方法。因此，我们旨在确定腕部佩戴的加速度计用于睡眠阶段分类的准确性，并随后描述睡眠时间和效率（卧床时总睡眠时间的比例）与死亡率结果之间的关联。
我们使用来自三个国家（澳大利亚、英国和美国）基于实验室的同步多导睡眠图和加速度计数据，开发并验证了一种用于睡眠阶段分类的自监督深度神经网络。该模型在队列内部使用受试者层面的五折交叉验证进行睡眠/清醒分类验证，并在三类设置（清醒、快速眼动睡眠（REM）、非快速眼动睡眠（NREM））下进行睡眠阶段分类验证，同时进行外部验证。我们通过将该模型应用于拥有10万名参与者的英国生物银行，评估了我们模型用于总体推断的表面效度，其中每位参与者佩戴腕带长达七天。导出的睡眠参数用于Cox回归模型，以研究睡眠时间和睡眠效率与全因死亡率之间的关联。
排除后，使用1448个参与者夜晚的数据来训练睡眠分类器。在外部验证中，多导睡眠图与模型分类之间的总睡眠时间差异为34.7分钟（95%一致性界限（LoA）：-37.8至107.2分钟），REM睡眠时间差异为2.6分钟（95% LoA：-68.4至73.4分钟），NREM睡眠时间差异为32.1分钟（95% LoA：-54.4至118.5分钟）。在英国生物银行样本中导出的睡眠结构估计显示出良好的表面效度。在66214名英国生物银行参与者中，观察到1642例死亡事件。与睡眠时间正常（6至7.9小时）的参与者相比，睡眠短者（<6小时）的死亡风险更高，无论他们的睡眠效率是低（风险比（HRs）：1.69；95%置信区间（CIs）：1.28至2.24）还是高（HRs：1.42；95% CIs：1.14至1.77）。
使用加速度计基于深度学习的睡眠分类与多导睡眠图有中等程度的一致性。我们的研究结果表明，夜间睡眠时间短会带来死亡风险，而与睡眠连续性无关。
