Abstract
With the emergence of machine learning for the classification of sleep and other human behaviors from accelerometer data, the need for correctly annotated data is higher than ever. We present and evaluate a novel method for the manual annotation of in-bed periods in accelerometer data using the open-source software Audacity®, and we compare the method to the EEG-based sleep monitoring device Zmachine® Insight+ and self-reported sleep diaries. For evaluating the manual annotation method, we calculated the inter- and intra-rater agreement and agreement with Zmachine and sleep diaries using interclass correlation coefficients and Bland–Altman analysis. Our results showed excellent inter- and intra-rater agreement and excellent agreement with Zmachine and sleep diaries. The Bland–Altman limits of agreement were generally around ±30 min for the comparison between the manual annotation and the Zmachine timestamps for the in-bed period. Moreover, the mean bias was minuscule. We conclude that the manual annotation method presented is a viable option for annotating in-bed periods in accelerometer data, which will further qualify datasets without labeling or sleep records.
Highlights
Utilizing machine learning with the identification of sleep, physical activity behavior, or non-wear from accelerometry data provides the ability to model very complex and non-linear relationships, which is not possible with more simple statistical methods, like multiple linear or logistic regression [1]
The algorithm in ZM has been compared to polysomnography (PSG) in adults with and without chronic sleep issues within a laboratory setting [23,24], and we found that ZM can be feasibly applied to children and adults for multiple days of measurements in free-living [12]
The Bland–Altman analysis revealed that the mean bias of the manual annotation and self-report compared to ZM was within ±6 min with LOA no larger than a span of ±45 min
Summary
Utilizing machine learning with the identification of sleep, physical activity behavior, or non-wear from accelerometry data provides the ability to model very complex and non-linear relationships, which is not possible with more simple statistical methods, like multiple linear or logistic regression [1]. Only scarce efforts to assess sleep measures from accelerometer data have been successfully approached with advanced machine learning techniques [7] providing researchers with inexpensive and minimally invasive methods. Valid objective measures using accurate automated scoring of sleep and wake time are important to provide valuable insights into the circadian rhythms on an individual and population-wide basis. The gold standard for objective sleep assessment is polysomnography (PSG), which is based on the continuous recording of electroencephalographic (EEG), electromyographic (EMG), and electrooculographic (EOG) activity via surface electrodes
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
