Multidimensional analysis of sleep electrophysiological signals
โ Scribed by D. Burger; P. Catani; J. West
- Book ID
- 104741635
- Publisher
- Springer-Verlag
- Year
- 1977
- Tongue
- English
- Weight
- 755 KB
- Volume
- 26
- Category
- Article
- ISSN
- 0340-1200
No coin nor oath required. For personal study only.
โฆ Synopsis
A night sleep polygraphic recording was coded by 20 s epochs, taking into account the recognizible features, summarized into 24 variables. In a 24 dimension space, the values of these variables are the coordinates of a point representative of each 20 s epoch. The coordinates of these points are given by a 24 column data table to which a factor analysis is applied in order to provide a bidimensional representation of the electrophysiological phenomena's evolution in time. Such a representation of the possible combinations of variables illustrates the structure of sleep without any biased ideas about this structure. The results obtained make it possible to consider the classical division of sleep into cycles and stages. Sleep seems to evolve continuously in time between a certain number of typical combinations, which supports the concept of cycle, but never following the same evolution between these combinations, which questions the reality of stages and their similarity from one cycle to the other.
The method described seems to reflect more accurately than the classical hypnogram the phenomena of continuous modulation of nervous activity during sleep, and to open interesting possibilities for an analytical automatic treatment of sleep electrophysiological signals.
๐ SIMILAR VOLUMES
The properties of two algorithms (zero-crossings algorithm and extrema algorithm) commonly applied to the automatic treatment of EEG signals are studied in full detail in order to make a comparison of their respective performances: quality of the signal restitution; reduction ratio; computation time
## Abstract Action potentials (APs) and field potentials (FPs) are commonly used to study functions of the brain. To facilitate their analysis, computer methods are often used. The highโspeed nature of APs and the large amount of data of FPs put stringent requirements on the sampling rate used. To