Sleep Stages Discrimination by Head Motion

In this study, sleep stages were estimated using features extracted from head acceleration. Two steps random forest was used for the estimation. As the result, it was possible to estimate sleep stages with more than 70% accuracy.