Preprint
Self-supervised learning of accelerometer data provides new insights for sleep and its association with mortality
The lancet
Elsevier Inc.
2023
Abstract
Background: 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.
Methods: 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.
Findings: 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).
Interpretation: 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.
Funding: This study is funded by National Health Service National Research Service, US National Institute of Health, Novo Nordisk, Health Data Research UK, Swiss Re, Wellcome Trust, the British Heart Foundation, the National Institute for Health Research, the UK National Institute for Health Research, the University of Western Australia, Curtin University, Telethon Kids Institute, Women and Infants Research Foundation, Edith Cowan University, Murdoch University, and the University of Notre Dame Australia and the Raine Medical Research Foundation.
Declaration of Interest: PG reports receiving funding from NIH/NIMH. AD receives funding from Wellcome Trust, Novo Nordisk, Swiss Re, the British Heart Foundation Centre of Research Excellence, and Health Data Research UK, an initiative funded by UK Research and Innovation, Department of Health and Social Care (England) and the devolved administrations, and leading medical research charities. ACR is supported by Flinders Foundation, Hospital Research Foundation, Compumedics, Vanda Pharmaceuticals, Teva Pharmaceuticals, and Sleep Health Foundation. DR receives funding from Wellcome Trust and Medical Research Council. DR also received lecture fees from Pfizer for an education program. KM is supported by Sir Charles Gairdner Hospital Research Advisory Council Funding, Early to Mid-Career Researchers Grant Medical Research Future Fund, Chevron Australia, CHC Helicopter Australia, Incannex Healthcare Limited, Nyxoah Pty Ltd. KM receives consulting fees from Melius Consulting, HIF, Invicta Medical and lecture fees from Sleep Health Foundation, WA Dental Association, and Shire of Cannington. Maja receives funding from the National Institute of Health.
Ethical Approval: This research has been conducted using the UK Biobank Resource under Application Number 59070. The UK Biobank received ethical approval from the National Health Service National Research Service (Ref 21/NW/0157).
Details
- Title
- Self-supervised learning of accelerometer data provides new insights for sleep and its association with mortality
- Authors/Creators
- Hang Yuan - University of OxfordTatiana Plekhanova - University of LeicesterRosemary Walmsley - University of OxfordAmy Reynolds - Flinders UniversityKathleen Maddison - The University of Western AustraliaMaja Bucan - University of PennsylvaniaPhilip Gehrman - University of PennsylvaniaAlex Rowlands - University of LeicesterDavid Ray - Oxford Centre for Diabetes, Endocrinology and MetabolismDerrick Bennett - University of OxfordJoanne McVeigh - Curtin UniversityLeon Straker - Curtin UniversityPeter EastwoodSimon KyleAiden Doherty - University of Oxford
- Publication Details
- The lancet
- Publisher
- Elsevier Inc.
- Identifiers
- 991005598470107891
- Murdoch Affiliation
- Health Futures Institute
- Resource Type
- Preprint
Metrics
4 File views/ downloads
95 Record Views