Automatic Detection of Sleep Stages Using Deep Learning Algorithm

Sleep plays a vital role in an individual's life, it enables proper cognitive and behavioral functions. An insufficient amount of sleep can lead to serious repercussions. Sleep disorders adversely affect a person's lifestyle and poses added risk of disorders such as diabetes, high blood pressure, heart disease and stroke, mood disorders, weight gain and obesity. Manual sleep staging and analysis is burdensome and time consuming, hence automatic sleep staging is required. Sleep is classified into various stages (Awake, N1, N2, N3 and REM). The aim of this paper is to survey the progress and challenges in various existing Electroencephalogram (EEG) signal-based methods used for sleep staging in an attempt to find the research gaps and possibly introduce a reasonable solution. Thus, in this paper, a novel and efficient technique to identify sleep stages using Convolutional Neural Network(CNN), CNN + LSTM (Long Short Term Memory) methods applied to 30s epochs of single-channel Fpz-Cz channels. In this study, the PhysioNet sleep cassette study data is used in European Data Format (EDF) Database was used. The proposed methodology achieves accuracy:84.19%, MF1:76.76% respectively.