Sleep Quality Differences According to a Statistical Continuous Sleep Model

  • A. G. Ravelo-García
  • F.D. Lorenzo-García
  • J.L. Navarro-Mesa
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 27)


This paper presents sleep quality differences between good and bad sleepers measured with a statistical continuous sleep model according to the Self-Rating Questionnaire for Sleep and Awakening Quality (SSA). Our main goal is to describe sleep continuous traces that take into account the sleep stage probability with a temporal resolution of 3 s, instead of the Rechtschaffen and Kales (R and K) resolution, which is 30 s. We adopt in our study the probability of being in stages W, S1, S2, S3, S4, and REM. The system uses only one electroencephalographic (EEG) channel. In order to achieve this goal we start by applying a hidden Markov model, in which the hidden states are associated with the sleep stages. These are probabilistic models that constitute the basis for the estimation of the sleep stage probabilities. The features that feed our model are based on the application of a discrete cosine transform to a vector of logarithmic energies at the output of a set of linearly spaced filters. In order to find differences between groups of sleepers, we define some measures based on the probabilistic traces. The experiments are performed over 24 recordings from the SIESTA database. The results show that our system performs well in finding differences in the presence of the Wake and S4 sleep stages for each group.


Hide Markov Model Sleep Quality Discrete Cosine Transform Sleep Stage Transition Probability Matrix 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.



We would like to thank Dr. Alpo Varri at Tampere University of Technology for providing us the database used in this research.


  1. 1.
    Rechtschaffen A, Kales A (1968) A manual of standardized terminology techniques and scoring system for sleep stages of human subjects. Brain Research Institute, UCLA, Los Angeles, USA. A threshold selection method from gray-level histograms. IEEE T Syst Man Cybernetics 9(1):62–66Google Scholar
  2. 2.
    Nielsen KD (1993) Computer assisted sleep analysis. Doctoral Thesis, Aalborg University, DenmarkGoogle Scholar
  3. 3.
    Navarro JL, Ravelo AG, Lorenzo FD, Martín SI, Hernández E, Quintana P (2006) On the determination of differences between good and bad sleepers by means of a hidden Markov model. WSEAS T Comput Res 1:321–324Google Scholar
  4. 4.
    Flexer A, Dorffner G, Sykacek P, Rezek I (2002) An automatic, continuous and probabilistic sleep stager based on a hidden Markov model. Appl Artif Intell 16(3):199–207CrossRefGoogle Scholar
  5. 5.
  6. 6.
    Rabiner LR, Juang BH (1993) Fundamentals of speech recognition. Prentice-Hall, Englewood Cliffs, NJGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2009

Authors and Affiliations

  • A. G. Ravelo-García
    • 1
  • F.D. Lorenzo-García
    • 1
  • J.L. Navarro-Mesa
    • 1
  1. 1.Dpto. de Señales y ComunicacionesUniversidad de Las Palmas de Gran CanariaE35017Spain

Personalised recommendations