Sleep Detection Using Physiological Signals from a Wearable Device

Sleep classification using Consumer Sleep Technologies and AI: A review of the current landscape

“…Others use hours or epochs to describe the data. Also as seen in Table 1 not every study uses PSG validated data [38]. as an example made subjects self report when they went to bed and when they were awake and used that as ground truth for awake/sleep classification.…”

“…This has been done in Ref. [38] where they use the Empatica E4 wrist device that has various sensors amongst which they use PPG, Electrodermal activity (EDA) and the accelerometer. They extract features for the sensors for a given 30 s segment of data like the mean of EDA, mean heart rate, standard deviation for the Blood Volume Pulse (BVP) and mean of the accelerometer signal.…”