"Sleep-in-a-box": Algorithmic approach to extract rest phases with mobile accelerometry
- Published
- Accepted
- Subject Areas
- Bioengineering, Bioinformatics, Anatomy and Physiology, Computational Science
- Keywords
- Mobile accelerometry, Sleep, Restphase, Main-Rest-Block
- Copyright
- © 2015 Müller et al.
- Licence
- This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ PrePrints) and either DOI or URL of the article must be cited.
- Cite this article
- 2015. "Sleep-in-a-box": Algorithmic approach to extract rest phases with mobile accelerometry. PeerJ PrePrints 3:e978v1 https://doi.org/10.7287/peerj.preprints.978v1
Abstract
This paper suggests a knew approach for analysing inactive phases in long term measurements using mobile accelerometry attached to a human body. The goal of the work is to provide a commandline tool to automaticaly parse a timestamped list of acceleration data in order to extract the phases that represent the sleeping periods. Justified on the fact that the data collection is motivated more in a healthcare environment than in a detailed sleeping research the methods do not concentrate on getting exact differentiation between real sleeping and awake lying periods. Detecting a significant long period with an inactivity pattern that permits the assumption that the received recovery quality is similar to the effects of sleeping on the human body will be enough to classify it as restphase. Furthermore the algorithm is not designed to find the exact amount of time that can be related to this classification. Finding enough of the relevant periods to provide a good view on the quantity and quality of inactive rest the observed human body gets should satisfy the task.
Author Comment
Our motivation of "Sleep-in-a-box-" measurement: The primary objective is observing the rest of a patient without influencing his average behavior. Therefore a long term measurement is applied to get information about daily rest rhythm. In this context maximizing adherence by using a small inconspicuous sensor and respecting privacy by abstracting sensible data is inevitable. Within a students work for the 2nd Winter Symposium of the Human Motion Project an algorithm was developed to enable easy comparison of different measurements by visualizing the time periods that are related to the Main-Rest-Phase. The focus was on developing a method suitable for automatic and structured offline analysis of huge data sets.