This tool converts actigraphy data into a prediction of DLMO time using mathematical models of the human
circadian clock. To try it, upload actigraphy files (in .csv format) using the button below. You can
upload multiple files at a time.
Only tested in Safari, Chrome, and Firefox.
Created by Philip Cheng, PhD and Olivia Walch, PhD
Research supporting this tool is funded by the National Heart Lung and Blood Institute (K23HL138166)
This website is a prototype that uses a mathematical model of the circadian clock to predict timing of DLMO from actigraphy data. The output of the model is hours after midnight (in the local timezone of the browser) on the last day in the actigraphy file. Please reach out to the authors if you have any questions.
You can download the template for your actigraphy data here and sample data here. We don't store any of the data you upload.
Important notes: Significant amounts of NaN data will affect the accuracy of the model as will periods where the watch was not worn at the beginning and ending of collection. Those epochs should be deleted prior to upload. Remember: "Garbage in, garbage out"!
For validation information, or to cite the use of this tool:
Cheng P, Walch O, Huang, Y., Mayer, C., Sagong, C., Cuamatzi Castelan, A., Burgess, HJ, Roth, T.,
Forger, D., Drake, CL. (in press). Predicting circadian misalignment with wearable technology: Validation of
wrist-worn actigraphy and photometry in night shift workers, SLEEP
Want to try an iOS implementation of DLMO prediction? A TestFlight build for APSS 2023 is available here.