Digital Technology for Sleep Symptoms in Parkinson's Disease: A Scoping Review.
This scoping review maps 19 studies using wearable, portable, and contactless digital technologies to assess diverse sleep disturbances in Parkinson's disease, classifying tools by technical maturity and highlighting limited disease-specific validation and small cohorts.
What the AI sees
This scoping review maps 19 studies using wearable, portable, and contactless digital technologies to assess diverse sleep disturbances in Parkinson's disease, classifying tools by technical maturity and highlighting limited disease-specific validation and small cohorts.
Research significance
By identifying emerging digital sleep biomarkers and gaps in multimodal, validated longitudinal monitoring, the paper points to opportunities to develop scalable endpoints and patient-centered phenotyping that could improve trial design and therapeutic evaluation in PD, though prospective…
Source abstract
Sleep disturbances are highly prevalent and clinically significant non-motor features of Parkinson's disease (PD). Although in-laboratory polysomnography remains the gold standard, its limited scalability and ecological validity constrain longitudinal and real-world assessment. Recent advances in digital health technologies have introduced a broad spectrum of portable, wearable, and contactless tools for sleep monitoring. In this scoping review, we systematically map the landscape of digital sleep technologies in PD using a tiered framework based on technical maturity and clinical validation (Tier 1-4), and further classify them by signal modality and sleep symptom domain. Through a systematic review of the literature, we identified 19 studies (Tier 2-4) applying digital biomarkers to assess sleep disturbances in PD, including REM sleep behavior disorder, nocturnal immobility, insomnia, circadian rhythm disturbances, excessive daytime sleepiness, and sleep-related respiratory and movement disorders. We additionally contextualize these findings against the rapid expansion of multimodal and AI-driven Tier 3-4 platforms in the general population. Despite this technological progress, a major translational gap persists in PD, characterized by limited disease-specific validation, small cohort sizes, and insufficient multimodal benchmarking. Multimodal systems leveraging machine learning offer a promising direction by enabling more precise characterization of complex and overlapping sleep phenotypes. Emerging contactless systems further expand the potential for continuous, low-burden monitoring, although their clinical validity remains to be established. Future development of digital sleep biomarkers in PD will require prospective validation against established standards and integration of multimodal data to enable scalable, longitudinal phenotyping and clinical trial applications.