AS3-SAE: Automatic Sleep Stages Scoring using Stacked Autoencoders

Abstract

Sleep is a subconscious state, and the brain is active during it. Automatic classification of sleep stages can help identify various diseases. In this paper, a deep learning type neural network called Stacked Autoencoders (SAEs) is used to automatically classify sleep stages with high computing speed, which is robust to noise. SAEs is a kind of neural networks with two encoder and decoder blocks, and ten hidden layers in each block. The function of these networks is similar to the human brain, and is capable of automatically processing signals. To prove the efficiency of this network, in addition to examining the effect of various biological signals such as ECG and EEG on the performance of sleep stage classification, SHHS and ISRUC standard databases have been used. The accuracy of classifying 2 to 6 classes by SHHS database are 1.00, 0.993, 0.9880, 0.9688, 0.961, and on ISRUC database accuracies are 1.00, 1.00, 0.996, 0.9431. Moreover, the proposed network can classify wake, deep sleep, and light sleep using the ECG signal (acc=0.75, kappa=0.69). In the review of the results, it is concluded that sleep stages classification based on EEG signal have better results, still acquisition of ECG signal, and its acceptable results can be a good alternative to use. In addition to its high ability of the proposed method to detect sleep stages, this network is robust to noise, which is very necessary and important for the clinical processing of sleep signals.

Keywords

• Sleep stages
• Stacked Autoencoder
• Single channel EEG
• Deep learning
• ECG.