Home-Based Detection of Isolated REM Sleep Behavior Disorder Using a Lumbar Wearable Sensor

Overview

Isolated REM Sleep Behavior Disorder (iRBD) can be sensitively detected at home using a lumbar-mounted wearable sensor over multiple nights. Machine learning analysis of mobility patterns from this sensor demonstrated high sensitivity and moderate specificity in distinguishing iRBD patients from controls, with performance improving up to five nights of data collection.

Background

iRBD is a strong predictor of neurodegenerative synucleinopathies and is traditionally diagnosed via overnight video-polysomnography (vPSG) in specialized sleep laboratories. However, access to vPSG is limited and sleep habits vary, complicating diagnosis. Wearable sensors offer a potential solution for home-based monitoring, enabling multi-night assessment of nocturnal motor activity associated with iRBD. This study evaluated the feasibility of using a lumbar-mounted inertial measurement unit to detect iRBD in a home setting.

Data Highlights

Key Findings

• Distinct nocturnal mobility patterns were observed in iRBD participants compared to controls using a lumbar wearable sensor.
• Machine learning models trained on mobility features achieved high sensitivity and moderate specificity for iRBD classification.
• Classification performance improved with the number of nights recorded, plateauing at five nights.
• Principal component analysis revealed significant differences between laboratory-based vPSG data and home-based wearable sensor data.
• Multi-night home monitoring with lumbar-mounted wearables is feasible and supports sensitive detection of iRBD motor patterns.

Clinical Implications

Lumbar-mounted wearable sensors provide a practical, non-invasive method for home-based screening of iRBD, potentially overcoming limitations of access to laboratory vPSG. Multi-night data collection enhances diagnostic accuracy, suggesting that a staged screening approach incorporating wearable technology could facilitate earlier identification and cohort enrichment for further neurodegenerative disease evaluation.

Conclusion

This study demonstrates that lumbar wearable sensors can sensitively detect iRBD-related nocturnal motor activity at home over multiple nights, offering a promising tool for accessible and scalable screening of this prodromal neurodegenerative condition.

References

1. Postuma et al. 2019 -- Risk and predictors of dementia and Parkinsonism in idiopathic REM sleep behaviour disorder
2. Cesari et al. 2022 -- Video-polysomnography procedures for diagnosis of rapid eye movement sleep behavior disorder
3. Stefani & Högl 2021 -- Sleep disorders in Parkinson's disease
4. Oz et al. 2023 -- Monitoring sleep stages with a soft electrode array
5. Cygan et al. 2010 -- Night-to-night variability of muscle tone, movements, and vocalizations in REM sleep behavior disorder