Mathematics > Dynamical Systems
[Submitted on 19 Jul 2022 (this version), latest version 13 Jun 2023 (v3)]
Title:A Level Set Kalman Filter Approach to Estimate the Circadian Phase and its Uncertainty from Wearable Data
View PDFAbstract:Daily (~24hr) rhythms of behavior and physiology such as sleep and hormone secretion are coordinated by an endogenous timer, the circadian clock. The accurate estimation of the clock state (i.e., the circadian phase) outside of the laboratory has enormous potential for precision medicine. Several methods that predict the phase from measurements collected with wearables (e.g., Apple Watch) have been recently developed. However, computation of the uncertainty in the estimation remains an open problem. The uncertainty analysis is necessary because the estimation accuracy can largely change even by a small perturbation of daily routine. Here, we present a method to account for the uncertainty and estimate the circadian phase using a new extension of Kalman filtering named the level set Kalman filter. Using the newly proposed method, we study the relationship between phase uncertainty and process noise from various sources. This allows the identification of the magnitude of the noise in the circadian system, which is impossible with previous methods. Moreover, our study reveals how much the uncertainty of the phase estimate of the central clock that is inaccessibly located in the brain can be reduced when measurements of the peripheral clock phase are given from wearables. We also show that our method has a performance improvement over the previous methods. Finally, we apply our method to real-world data to further identify its usefulness. These results set the stage for systematically understanding the circadian dynamics in the real world.
Submission history
From: Dae Wook Kim [view email][v1] Tue, 19 Jul 2022 17:00:08 UTC (1,114 KB)
[v2] Sat, 10 Jun 2023 08:32:49 UTC (1,123 KB)
[v3] Tue, 13 Jun 2023 06:02:29 UTC (2,240 KB)
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.