Monitoring sleep stages with a soft electrode array: Comparison against<scp>vPSG</scp>and home‐based detection of<scp>REM</scp>sleep without atonia

Sleep disorders are symptomatic hallmarks of a variety of medical conditions. Accurately identifying the specific stage in which these disorders occur is particularly important for the correct diagnosis of non-rapid eye movement and rapid eye movement parasomnias. In-lab polysomnography suffers from limited availability and does not reflect habitual sleep conditions, which is especially important in older adults and those with neurodegenerative diseases. We aimed to explore the feasibility and validity of a new wearable system for accurately measuring sleep at home. The system core technology is soft, printed dry electrode arrays and a miniature data acquisition unit with a cloud-based data storage for offline analysis. The positions of the electrodes allow manual scoring following the American Association of Sleep Medicine guidelines. Fifty participants (21 healthy subjects, mean age 56.6 ± 8.4 years; and 29 patients with Parkinson's disease, 65.4 ± 7.6 years) underwent a polysomnography evaluation with parallel recording with the wearable system. Total agreement between the two systems reached Cohen's kappa (k) of 0.688 with agreement in each stage of: wake k = 0.701; N1 = 0.224; N2 = 0.584; N3 = 0.410; and rapid eye movement = 0.723. Moreover, the system reliably detected rapid eye movement sleep without atonia with a sensitivity of 85.7%. Additionally, a comparison between sleep as measured in the sleep lab with data collected from a night at home showed significantly lower wake after sleep onset at home. The results demonstrate that the system is valid, accurate and allows for the exploration of sleep at home. This new system offers an opportunity to help detect sleep disorders on a larger scale than possible today, fostering better care.