End to End Sleep Staging using Fiber Optic Signal

Sleep is an essential process in everyone's life, and sleep quality is closely related to human physical and mental health. For checking the quality of human's sleep, going through the sleep staging process would be the first step to take. In this paper, an end-to-end network is introduced to extract the living sign features from the optical fiber signal. Experiments show that fiber optic signal can replace PSG(Polysomnography) for sleep staging to some extent, and the deep learning model-based non-contact device is as accurate as the traditional EEG device. Finally, the detected sleep staging is used to calculate the sleep score and efficiency to evaluate the quality of the whole night sleep.