Classification of Sleep Videos Using Deep Learning

摘要

Videosomnography (VSG) is a group of video-based methods used to record and label sleep versus awake states in humans. Traditional behavioral-VSG (B-VSG) labeling requires visual inspection of the video by a trained technician to determine whether a subject is sleep or awake. B-VSG is not used to label sleep stages (e.g., slow wave or REM sleep), rather it solely labels whether a subject is asleep or awake at a particular time. In this paper we describe an automated VSG sleep detection system which uses deep learning approaches to label frames in a sleep video as "sleep" or "awake" in young children. We examine 3D Convolutional Networks (C3D) and Long Short-term Memory (LSTM) relative to motion information from selected Groups of Pictures of a sleep video and test temporal window sizes for back propagation. We compared our proposed VSG methods to traditional B-VSG sleep-awake labels. C3D had an accuracy of approximately 90% and the proposed LSTM method improved the accuracy to more than 95%. The analyses revealed that estimates generated from the proposed LSTM-based method with long-term temporal dependency are suitable for automated sleep or awake labeling.