An Ample Review of Various Deep Learning Skills for Identifying the Stages of Sleep

Sleep is an important part of everyone’s life. Sleep disorders are very common nowadays, when neglected may cause many neurological problems. Common problems like sleep interruptions, snoring, etc. can be detected by using sleep stage analysis. However, the conventional methods used for sleep analysis are very time-consuming. This limitation can be avoided by using an automatic tool for diagnosis based on artificial intelligence. Different artificial intelligence technologies like deep learning ensure the full utilization of data with very less information loss. A detailed study of different models from the years is provided here. The studies here use deep learning model techniques to analyze the sleep stages using polysonogram (PSG) signals. Additionally, this investigation demonstrates the use of electrocardiogram (ECG) signals with convolutional neural networks for the analysis of the various stages of sleep. It demonstrates the excellent categorization performance of CNN networks, especially 1D CNN. An examination of the studies reveals that EOG and EMG signals may also be utilised in future automated detection systems, in addition to EEG signals. This led us to the conclusion that PSG signals can also be used in conjunction with deep learning algorithms in addition to EEG signals. This study examines various techniques using PSG signals and deep learning skills that have been published in latest years.