Deep Learning for Sleep Staging Using EEG Signals

Sleep plays a crucial role in the physical and mental health of an individual. The importance of sleep in the well-being of an individual is widely recognized, and so are the effects of the sleep disorders or disturbances. These sleep disorders, for example insomnia or sleep apnea, affect the individual's sleep structure and quality, and are worrisomely prevalent across all age and sex categories. There exists a substantial body of literature which indicates that sleep disturbances can severely reduce the quality of life and, additionally, present themselves as risk factors in the development of other health or psychological disorders, such as cardiovascular disease or depression. Sleep staging, the classification of the stages of sleep, is a valuable tool in the detection of sleep disorders such as insomnia. This paper proposes a Deep Learning-based method for sleep staging, using single-channel EEG data. A CNN-LSTM Deep Neural Network is used. The paper uses the Expanded Sleep EDF public dataset.