Sleep Stages Classification Using Neural Networks with Multi-channel Neural Data

Recent studies on sleep reveal its importance not only on body functioning but also in brain and cognition health. As the development of wearable smart devices, neural data such as electroencephalogram (EEG) becomes commonly accessible. Developing good algorithm for sleep classification is more urgent and necessary than before. This paper presents a new and robust 5-stage sleep classification algorithm based on feedforward neural network, with flexible model structure using normalized Power Spectral Density (PSD) collected from multi-channel neural data. The algorithm is described with detailed mathematical elaboration. It utilizes the power of deep learning and provides average accuracy close to 90% which is competitive, compared with previous researches.