due to shift work) have been understudied. The result is that unconventional sleep patterns (eg. In addition, actigraphy devices do not provide real-time feedback about the user’s sleep, which hinders longitudinal monitoring. This limitation extended to proprietary algorithms used by commercial wearables. Using them with data recorded over the entire day is unreliable, even with sleep annotations provided by PSG technicians or the subjects themselves 29, 30. Additionally, actigraphy algorithms have only been optimised to detect sleep at night, as study participants typically spend one or two nights in the laboratory. Actigraphy often struggles to classify wake events during the sleep period, yielding poor specificity compared to PSG 16, 21, 23, 27, 28. These algorithms have been widely adopted and were recently bench-marked against both each other and newer machine learning and deep learning methods 23– 26.
Some have proved to have strong validity and reliability against PSG 17– 22. Over the past 30 years, several actigraphy-based algorithms have been developed to detect night-time sleep and wake periods. At present, the use of actigraphy in healthy sleepers is approved by both the FDA and recommended by the American Academy of Sleep Medicine (AASM) 14. The advantage over PSG is that actigraphy, and modern counterpart, accelerometry, require sensors amenable to affordable wrist-worn devices 13, 16. Since then, many studies have assessed actigraphy against PSG 13– 15. It originates in early telemetric measurements of motor activity in the 1970s which were used to assess sleep quality 12. Furthermore, the unfamiliar laboratory environment might not favor the patients’ typical sleep pattern 11.Īctigraphy is an established and popular alternative to PSG. This has limited large-scale and long-term population studies. PSG requires signals from multiple sensors, as well as expert input and is thus limited to the laboratory. The gold-standard method for quantifying sleep quantity and quality is polysomnography (PSG). Apart from outright sleep disorders, sleep patterns impact the quality of life and history of common diseases, whether cardiovascular, metabolic or neurodegenerative.
This makes the objective monitoring of sleep crucial for understanding human health. As such, sleep lies at the cross-roads of multiple research programs in both life sciences and public health. Its functions are not fully understood, despite extensive studies on its its influence on energy homeostasis, immune function, cognitive performance and behaviour 2– 10. Human sleep is a reversible physiological state that is essential for health and performance 1.
0 kommentar(er)
