Please use this identifier to cite or link to this item:
|Scopus||Web of Science®||Altmetric|
|Title:||A new quantitative automatic method for the measurement of non-rapid eye movement sleep electroencephalographic amplitude variability|
|Citation:||Journal of Sleep Research, 2012; 21(2):212-220|
|Raffaele Ferri, Francesco Rundo, Luana Novelli, Mario G. Terzano, Liborio Parrino and Oliviero Bruni|
|Abstract:||The aim of this study was to arrange an automatic quantitative measure of the electroencephalographic (EEG) signal amplitude variability during non-rapid eye movement (NREM) sleep, correlated with the visually extracted cyclic alternating pattern (CAP) parameters. Ninety-eight polysomnographic EEG recordings of normal controls were used. A new algorithm based on the analysis of the EEG amplitude variability during NREM sleep was designed and applied to all recordings, which were also scored visually for CAP. All measurements obtained with the new algorithm correlated positively with corresponding CAP parameters. In particular, total CAP time correlated with total NREM variability time (r = 0.596; P < 1E-07), light sleep CAP time with light sleep variability time (r = 0.597; P < 1E-07) and slow wave sleep CAP time with slow wave sleep variability time (r = 0.809; P < 1E-07). Only the duration of CAP A phases showed a low correlation with the duration of variability events. Finally, the age-related modifications of CAP time and of NREM variability time were found to be very similar. The new method for the automatic analysis of NREM sleep amplitude variability presented here correlates significantly with visual CAP parameters; its application requires a minimum work time, compared to CAP analysis, and might be used in large studies involving numerous recordings in which NREM sleep EEG amplitude variability needs to be assessed.|
|Keywords:||automatic analysis; cyclic alternating pattern; NREM sleep variability; sleep scoring|
|Rights:||© 2011 European Sleep Research Society|
|Appears in Collections:||Medicine publications|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.