Title: AUTOMATIC SLEEP/WAKE STAGING OF RAT POLYSOMNOGRAPHIC RECORDINGS USING TWO-STEP SYSTEM

Issue Number: Vol. 1, No. 2
Year of Publication: 2011
Page Numbers: 415-425
Authors: Lukas Zoubek, Florian Chapotot
Journal Name: International Journal of Digital Information and Wireless Communications (IJDIWC)
- Hong Kong

Abstract:


This article describes application of an automatic classification system on a database of rat polysomnographic recordings. A two-step automatic system is used to analyze and score the recordings into three stages: wake, NREM sleep and REM sleep. First step of the analysis consists in artifact identification which leads to the selection of artifact-free signals. Then, 30sec epochs are classified according to relevant features extracted from the available artifact-free signals using artificial neural network classifiers. The overall classification accuracy reached by the presented classification system exceeded 95%, when analyzed 53 polysomnographic recordings.