Real-time home REM sleep detection using Apple watch, FITBIT, and Cerebra
LE3 .A278 2023
2023
Leslie, Kenneth
Acadia University
Bachelor of Science
Honours
Psychology
As smartwatches increase in popularity, new opportunities arise for their further integration into everyday life, such as the detection and influence of dreaming. The present study uses the Cerebra home PSG system as the ground truth for detecting sleep stages. Data was also recorded using an Apple Watch and a Fitbit. The Apple Watch data was used to develop a model capable of real-time REM sleep detection, created by the Acadia Institute for Data Analytics (AIDA). The Fitbit’s accuracy at detecting sleep metrics was assessed, to determine how accurate a mature sleep staging algorithm can be based on heart rate and movement data. Participants (N=4) completed a minimum of six nights of sleep recording and were compensated via $60 Amazon gift cards. The Fitbit detected true positive REM periods (Fitbit and Cerebra agreement) at a rate of 85.1%. Various tests were conducted to determine if there were any differences in the sleep metrics recorded by the Fitbit and the Cerebra, and no significant differences were found. The findings show promise for the future new dreaming technology using wearable smartwatches.
The author retains copyright in this thesis. Any substantial copying or any other actions that exceed fair dealing or other exceptions in the Copyright Act require the permission of the author.
https://scholar.acadiau.ca/islandora/object/theses:4025