Activity-based sleep-wake identification in infants

Actigraphy offers one of the best known alternatives to polysomnography for wake-sleep identification. The advantages of actigraphy include high accuracy, simplicity of use and low intrusiveness. These features allow of use actigraphy for determining wake-sleep states in highly sensitive groups such as infants. Both logistic regression and neural networks were tested as predictors. The accuracy of predicted wake-sleep states were established in comparison to the sleep/wake states recorded by technicians in a polysomnograph study. Both prediction methods provided good accuracy of prediction and agreement in results with other studies, thus validating the suggested methodology.