Automated identification of sleep states from EEG signals by means of ensemble empirical mode decomposition and random under sampling boosting

In clinical practice, sleep stage annotation is typically performed by an expert scorer on the basis of visual examination of polysomnographic (PSG) measurements which is composed of electroencephalogram (EEG), electromyogram (EMG) and electrooculogram (EOG). In this respect, Rechtschaffen ’s and Kales’s (R&K) recommendations  [1] are widely followed. The annotation of 8-hour (whole night) recording requires approximately 2-4 hours which is not pragmatic in current clinical settings [2]. Moreover, visual inspection of this gargantuan volume of data not only makes this process onerous for clinicians but also makes sleep scoring subject to human error, monotonous and dependent on expensive human resources.
Source: Computer Methods and Programs in Biomedicine - Category: Bioinformatics Authors: Source Type: research