Dorothée Coppieters 't Wallant1, Vincenzo Muto2, Giulia Gaggioni3, Mathieu Jaspar4, Sarah L Chellappa5, Christelle Meyer6, Gilles Vandewalle7, Pierre Maquet8, Christophe Phillips9. 1. Cyclotron Research Centre, University of Liège, Allée du 6 Août 8 B30, B-4000 Sart-Tilman, Belgium; Department of Electrical Engineering and Computer Science, University of Liège, Allée de la découverte 10 B28, B-4000 Liège, Belgium. Electronic address: d.coppieters@ulg.ac.be. 2. Cyclotron Research Centre, University of Liège, Allée du 6 Août 8 B30, B-4000 Sart-Tilman, Belgium; Department of Psychology: Cognition and Behaviour, University of Liège, Place des Orateurs 2, B32B-4000 Liège, Belgium; Walloon Excellence in Lifesciences and Biotechnology (WELBIO), Avenue Pasteur 6, B-1300 Wavre, Belgium. Electronic address: vincenzo.muto@ulg.ac.be. 3. Cyclotron Research Centre, University of Liège, Allée du 6 Août 8 B30, B-4000 Sart-Tilman, Belgium. Electronic address: giulia.gaggioni@ulg.ac.be. 4. Cyclotron Research Centre, University of Liège, Allée du 6 Août 8 B30, B-4000 Sart-Tilman, Belgium; Department of Psychology: Cognition and Behaviour, University of Liège, Place des Orateurs 2, B32B-4000 Liège, Belgium; Walloon Excellence in Lifesciences and Biotechnology (WELBIO), Avenue Pasteur 6, B-1300 Wavre, Belgium. Electronic address: mathieu.jaspar@ulg.ac.be. 5. Cyclotron Research Centre, University of Liège, Allée du 6 Août 8 B30, B-4000 Sart-Tilman, Belgium. Electronic address: sarah.chellappa@gmail.com. 6. Cyclotron Research Centre, University of Liège, Allée du 6 Août 8 B30, B-4000 Sart-Tilman, Belgium; Walloon Excellence in Lifesciences and Biotechnology (WELBIO), Avenue Pasteur 6, B-1300 Wavre, Belgium. Electronic address: christelle.meyer@ulg.ac.be. 7. Cyclotron Research Centre, University of Liège, Allée du 6 Août 8 B30, B-4000 Sart-Tilman, Belgium. Electronic address: Gilles.Vandewalle@ulg.ac.be. 8. Cyclotron Research Centre, University of Liège, Allée du 6 Août 8 B30, B-4000 Sart-Tilman, Belgium; Department of Neurology, University of Liège Hospital, B35, B-4000 Liège, Belgium; Walloon Excellence in Lifesciences and Biotechnology (WELBIO), Avenue Pasteur 6, B-1300 Wavre, Belgium. Electronic address: pmaquet@ulg.ac.be. 9. Cyclotron Research Centre, University of Liège, Allée du 6 Août 8 B30, B-4000 Sart-Tilman, Belgium; Department of Electrical Engineering and Computer Science, University of Liège, Allée de la découverte 10 B28, B-4000 Liège, Belgium. Electronic address: c.phillips@ulg.ac.be.
Abstract
BACKGROUND: In sleep electroencephalographic (EEG) signals, artifacts and arousals marking are usually part of the processing. This visual inspection by a human expert has two main drawbacks: it is very time consuming and subjective. NEW METHOD: To detect artifacts and arousals in a reliable, systematic and reproducible automatic way, we developed an automatic detection based on time and frequency analysis with adapted thresholds derived from data themselves. RESULTS: The automatic detection performance is assessed using 5 statistic parameters, on 60 whole night sleep recordings coming from 35 healthy volunteers (male and female) aged between 19 and 26. The proposed approach proves its robustness against inter- and intra-, subjects and raters' scorings, variability. The agreement with human raters is rated overall from substantial to excellent and provides a significantly more reliable method than between human raters. COMPARISON: Existing methods detect only specific artifacts or only arousals, and/or these methods are validated on short episodes of sleep recordings, making it difficult to compare with our whole night results. CONCLUSION: The method works on a whole night recording and is fully automatic, reproducible, and reliable. Furthermore the implementation of the method will be made available online as open source code.
BACKGROUND: In sleep electroencephalographic (EEG) signals, artifacts and arousals marking are usually part of the processing. This visual inspection by a human expert has two main drawbacks: it is very time consuming and subjective. NEW METHOD: To detect artifacts and arousals in a reliable, systematic and reproducible automatic way, we developed an automatic detection based on time and frequency analysis with adapted thresholds derived from data themselves. RESULTS: The automatic detection performance is assessed using 5 statistic parameters, on 60 whole night sleep recordings coming from 35 healthy volunteers (male and female) aged between 19 and 26. The proposed approach proves its robustness against inter- and intra-, subjects and raters' scorings, variability. The agreement with human raters is rated overall from substantial to excellent and provides a significantly more reliable method than between human raters. COMPARISON: Existing methods detect only specific artifacts or only arousals, and/or these methods are validated on short episodes of sleep recordings, making it difficult to compare with our whole night results. CONCLUSION: The method works on a whole night recording and is fully automatic, reproducible, and reliable. Furthermore the implementation of the method will be made available online as open source code.