Article Preview
TopIntroduction
Electroencephalogram (EEG) signals are considered as the standard indicators for classifying sleep stages. Sleep stages are classified as: Wake, Non-REM sleep (which is further sub-classified into stages N1, N2 and N3) and REM sleep. It is a well-known fact that the transition between wake and sleep stage 1 is influenced by wake - active neurons in multiple arousal centers and sleep-active neurons in the pre-optic area of the hypothalamus (Merica & Fortune (2004); Saper et al., (2001)). In principle, sleep onset separates these two behavioral states. The recognition of transition between wake and sleep stage 1 has a huge impact on the prevention of fatal accidents in critical professions like drivers, pilots, defense personnel and air traffic control operators (Z.Mardi et al.,(2011); Yilldiz et al., (2008); Lin et al.,(2005) Chung et al., (2005)).Although, automated sleep scoring techniques were being reported in the literature(Lainscsek et al., (2014); Virkkalaa et al., (2007); Liang et al., (2012);Gunes et al., (2010)), there is still a huge demand in terms of computational algorithm for appropriate recognition of transitions between wake and sleep stage1.This research work proposes a comparative study on the application of entropy feature such as fuzzy entropy with a time domain feature, relative spike amplitude (RSA) in order to determine the transition between wake and sleep stage 1. Figure 1 shows the proposed schematic diagram.
Figure 1. Proposed FE-RSA features based wake-sleep EEGs recognition
TopBrief Literature
Several attempts have been made to study the regional changes in cortical areas during wake-sleep transition. The factors that contribute to topographical changes and temporal changes during wake–sleep transition includes definition of sleep onset, frequency specific topographical changes, subject variability in terms of length of this transition(Rechtschaffen& Kales, 1968;Hasan& Broughton 1994;De Gennaro et al,2000).
Gennaro et al 2001 made a detailed study on the anterior – posterior changes during wake-sleep transition and concluded that the alpha rhythm spreads accurately as the transition progresses. The authors also made a detailed study to assess the relationship between slow eye movement and quantitative EEG measures during wake-sleep transition (Gennaro et al 2000).Sriraam et al (2014)have made use of relative spike amplitude and Hurst exponent features for detecting the transition between wake and sleep stage 1. In order to understand the time varying nature of wake-sleep stage 1 in EEG, Purnima et al; (2014) attempted by making use of spike rhythmicity feature which showed a significant discriminative feature for recognizing wake and sleep stage pattern.