Integrated analysis on the state of sleep through Polysomnography is crucial for diagnosis for sleep related disease. But conventional analog-type Polysomnography

Sleep stage scoring is laborious and requires much expertise. So using expert system for automatic sleep scoring systems is desirable. In this study, we have designed and implemented a multi-staged rule-based reasoning system with augmented functions using communications among RBRU(rulebased reasoning unit). With RBRU not having sufficient information, the unit is designed to ask the previous RBRU and SPU(signal processing unit) to provide more information. With brief information only, the RBRU can score sleep stage with efficiency and exactness in simple cases. If this brief information is not sufficient, additional information tossed from previous unit is combined for enhancing accuracy of sleep stage scoring.