Sleep Stages Classification Using Spectral Based Statistical Moments as Features

In the pursuit of portable, efficient and effective sleep staging systems, researchers have been testing a massive number of combinations of EEG features and classifiers. State of the art sleep classification ensembles achieve accuracy in the order of 90%. However, there is presently no consensus regarding the best set of features for identifying sleep stages with a single EEG channel, leading researchers to modify the feature selection according to the number of classification stages. This paper introduces a reduced set of frequency-domain features capable of yielding high classification accuracy (90.9%, 91.8%, 92.4%, 94.3% and 97.1%) for all 6to 2-state sleep stages. The proposed system uses fast Fourier transform (FFT) to convert data from Pz-Oz EEG channel into the frequency domain. Afterwards, eight statistical features are extracted from specific frequency ranges associated to brain rhythms, feeding a random forest classifier.